Latest version changes for Probe for EPMA (and CalcImage) v. 13.8.3

[John Donovan]

Just FYI, I added Mo to the geological sorted output as requested by Paul Carpenter.

Note that the "sort" order for analysis output can be "geological" (DHZ) order or ascending or descending atomic number order as selected in the Analytical |Analysis Options dialog as seen here:



The geological sort order is currently as follows (missing elements are added after the last one specified):

Code: [Select]

GeologicalOrder$(1) = "nb"
GeologicalOrder$(2) = "ta"
GeologicalOrder$(3) = "mo"
GeologicalOrder$(4) = "w"

GeologicalOrder$(5) = "si"
GeologicalOrder$(6) = "ge"
GeologicalOrder$(7) = "zr"
GeologicalOrder$(8) = "hf"
GeologicalOrder$(9) = "ti"
GeologicalOrder$(10) = "sn"
GeologicalOrder$(11) = "zn"
GeologicalOrder$(12) = "cd"
GeologicalOrder$(13) = "hg"
GeologicalOrder$(14) = "tl"
GeologicalOrder$(15) = "pb"
GeologicalOrder$(16) = "th"
GeologicalOrder$(17) = "u"

GeologicalOrder$(18) = "b"
GeologicalOrder$(19) = "al"
GeologicalOrder$(20) = "ga"
GeologicalOrder$(21) = "in"
GeologicalOrder$(22) = "v"
GeologicalOrder$(23) = "cr"

GeologicalOrder$(24) = "sc"
GeologicalOrder$(25) = "y"
GeologicalOrder$(26) = "la"
GeologicalOrder$(27) = "ce"
GeologicalOrder$(28) = "pr"
GeologicalOrder$(29) = "nd"
GeologicalOrder$(30) = "pm"
GeologicalOrder$(31) = "sm"
GeologicalOrder$(32) = "eu"
GeologicalOrder$(33) = "gd"
GeologicalOrder$(34) = "tb"
GeologicalOrder$(35) = "dy"
GeologicalOrder$(36) = "ho"
GeologicalOrder$(37) = "er"
GeologicalOrder$(38) = "tm"
GeologicalOrder$(39) = "yb"
GeologicalOrder$(40) = "lu"
GeologicalOrder$(41) = "ac"

GeologicalOrder$(42) = "fe"
GeologicalOrder$(43) = "co"
GeologicalOrder$(44) = "ni"
GeologicalOrder$(45) = "cu"
GeologicalOrder$(46) = "mn"

GeologicalOrder$(47) = "be"
GeologicalOrder$(48) = "mg"
GeologicalOrder$(49) = "ca"
GeologicalOrder$(50) = "sr"
GeologicalOrder$(51) = "ba"
GeologicalOrder$(52) = "ra"

GeologicalOrder$(53) = "li"
GeologicalOrder$(54) = "na"
GeologicalOrder$(55) = "k"
GeologicalOrder$(56) = "rb"
GeologicalOrder$(57) = "cs"
GeologicalOrder$(58) = "fr"

GeologicalOrder$(59) = "p"
GeologicalOrder$(60) = "s"
GeologicalOrder$(61) = "as"
GeologicalOrder$(62) = "se"
GeologicalOrder$(63) = "sb"
GeologicalOrder$(64) = "te"

GeologicalOrder$(65) = "he"
GeologicalOrder$(66) = "ne"
GeologicalOrder$(67) = "ar"
GeologicalOrder$(68) = "kr"
GeologicalOrder$(69) = "xe"
GeologicalOrder$(70) = "rn"

GeologicalOrder$(71) = "i"
GeologicalOrder$(72) = "br"
GeologicalOrder$(73) = "cl"
GeologicalOrder$(74) = "f"

GeologicalOrder$(75) = "c"
GeologicalOrder$(76) = "n"
GeologicalOrder$(77) = "o"
GeologicalOrder$(78) = "h"

[John Donovan]

One nice new change is a modified "Calibration Curve" dialog under the Run menu so even when not in Calibration Curve mode, the application will temporarily select Calibration Curve mode:

Warning in CalibrationLoad: Calibration curve mode not selected in the Analytical | Select ZAF, Phi-Rho-Z, Alpha Factor or Calibration Curve menu dialog.

and re-calculate the data for each sample and plotting it up in the calibration curve graph.

Note that the software distinguishes between off-peak and MAN (on-peak) measurements and creates a different calibration curve for each intensity type. Here one can see the Fayalite standard plotted up as an unknown in the Calibration curve dialog:



Although, this "Calibration Curve" quant method was developed specifically for carbon in steel measurements, the results on "normal" analyses is often surprisingly good, though not "robust" as they say. The quant results for the above Fayalite sample are shown here:

St  263 Set   1 Fe2SiO4 (synthetic fayalite), Results in Elemental Weight Percents
 
ELEM:       Si      Al      Fe      Mg      Cr      Ti      Mn      Ca       O
TYPE:     ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    ANAL    SPEC
BGDS:      LIN     LIN     LIN     LIN     LIN     LIN     LIN     LIN
TIME:    20.00   20.00   20.00   20.00   20.00   20.00   20.00   20.00
BEAM:    20.09   20.09   20.09   20.09   20.09   20.09   20.09   20.09

ELEM:       Si      Al      Fe      Mg      Cr      Ti      Mn      Ca       O   SUM 
   131  13.759   -.050  53.957    .549   -.016   -.016   -.008    .014  31.407  99.596
   132  13.840   -.061  53.898    .551    .007   -.012    .042    .046  31.407  99.717
   133  13.784   -.053  54.145    .550    .023    .009   -.034    .035  31.407  99.866
   134  13.811   -.053  53.973    .556   -.009    .043   -.031    .042  31.407  99.740
   135  13.815   -.062  54.092    .553   -.019    .033   -.002    .050  31.407  99.866

AVER:   13.802   -.056  54.013    .552   -.003    .011   -.007    .038  31.407  99.757
SDEV:     .031    .005    .102    .003    .017    .026    .030    .014    .000    .114
SERR:     .014    .002    .046    .001    .008    .012    .014    .006    .000
%RSD:      .22   -9.62     .19     .51 -594.05  230.03 -462.80   38.15     .00

PUBL:   13.785    n.a.  54.809    n.a.    n.a.    n.a.    n.a.    n.a.  31.407 100.001
%VAR:      .12     ---   -1.45     ---     ---     ---     ---     ---     .00
DIFF:     .017     ---   -.796     ---     ---     ---     ---     ---    .000
STDS:       14      13     395      12      24      22      25     358       0

Note that the standard for Si was SiO2 (std #14) and the standard for Fe was magnetite (#395) for the above analysis.

[Probeman]

Recently I discovered that the particle (and thin film) geometry correction method had been broken, but it is now fixed. Here is an example of a CaF2 1 um particle run first without the particle geometry correction as seen here:

CaF2 1 um particle- use Tetragonal Prism model

STANDARD PARAMETERS:

 ELEMENT  STDNUM STDCONC STDKFAC   Z-BAR  ABSCOR  FLUCOR  ZEDCOR  ZAFCOR
   Ca Ka     831  51.200   .5068 14.6320  1.0043  1.0000  1.0059  1.0102
    F Ka     831  48.800   .1080 14.6320  4.6000   .9998   .9822  4.5173

 ELEMENT STP-POW BKS-COR   F(x)e   F(x)s      Eo      Ec   Eo/Ec
   Ca Ka  1.0468   .9609   .9383   .9343   20.00  4.0390  4.9517
    F Ka   .9498  1.0341   .6357   .1382   20.00   .6870 29.1121

SAMPLE: 5, ITERATIONS: 12, Z-BAR: 12.61798

 ELEMENT  ABSCOR  FLUCOR  ZEDCOR  ZAFCOR STP-POW BKS-COR   F(x)u      Ec   Eo/Ec    MACs
   Ca ka  1.0060  1.0000  1.0096  1.0157  1.0644   .9485   .9326  4.0390  4.9517 148.028
   F  ka  3.1457   .9999   .9882  3.1082   .9678  1.0211   .2021   .6870 29.1121 4483.58

 ELEMENT   K-RAW K-VALUE ELEMWT% OXIDWT% ATOMIC% FORMULA KILOVOL             
   Ca ka  .60920  .30876  31.361   -----  18.852   8.000   20.00
   F  ka 1.90570  .20587  63.988   -----  81.148  34.436   20.00
   TOTAL:                 95.349   ----- 100.000  42.436


Note that although the total is fairly good (~95%), the atomic ratio (19:81) for Ca and F is very far off from what we should expect, namely 33:66.

Here is the same data now calculated with the particle geometry correction:

CaF2 1 um particle- use Tetragonal Prism model

STANDARD PARAMETERS:

 ELEMENT  STDNUM STDCONC STDKFAC   Z-BAR  ABSCOR  FLUCOR  ZEDCOR  ZAFCOR
   Ca Ka     831  51.200   .5069 14.6320  1.0043  1.0000  1.0059  1.0102
    F Ka     831  48.800   .1082 14.6320  4.5945   .9998   .9822  4.5119

 ELEMENT STP-POW BKS-COR   F(x)e   F(x)s      Eo      Ec   Eo/Ec
   Ca Ka  1.0468   .9609   .9384   .9344   20.00  4.0390  4.9517
    F Ka   .9498  1.0341   .6374   .1387   20.00   .6870 29.1121

SAMPLE: 5, ITERATIONS: 5, Z-BAR: 14.64359

E-RANGE:  11.1913, INTE-STEP:  113
Particle or thin film corrections utilized were Tetragonal Prism (flat top and curved sides)

Particle parameters were a particle diameter of 1 microns, a particle density of 3.1 gm/cm^3, a thickness factor of 1, and a numerical integration step size of 0.00001 microns.

 ELEMENT  ABSCOR  FLUCOR  ZEDCOR  ZAFCOR STP-POW BKS-COR   F(x)u      Ec   Eo/Ec    MACs
   Ca ka  1.5595  1.0000  1.0059  1.5687  1.0467   .9610   .6017  4.0390  4.9517 145.163
   F  ka  2.2710   .9999   .9822  2.2303   .9497  1.0342   .2807   .6870 29.1121 6437.34

 ELEMENT   K-RAW K-VALUE ELEMWT% OXIDWT% ATOMIC% FORMULA KILOVOL NORMEL% NORMOX%
   Ca ka  .60920  .30877  48.436   -----  33.307   8.000   20.00 51.3053   -----   
   F  ka 1.90570  .20612  45.971   -----  66.693  16.019   20.00 48.6947   -----
   TOTAL:                 94.408   ----- 100.000  24.019         100.000   -----


Note that now the Ca:F atomic ratio is now as it should be (33:66) even though the total is actually slightly worse!

See this topic for more information on particle corrections:

https://probesoftware.com/smf/index.php?topic=281.0

[John Donovan]

One can now edit the sample name or title for CalcImage output as seen here:



Idea from Paul Carpenter!  Thanks, Paul!
john

[John Donovan]

New "interactive Help" buttons:

[John Donovan]

The latest version (10.4.2) has a new feature that allows one to search the standard database for compositions with large matrix corrections.

From the Options | Find menu this dialog is available:



Based on the standards found based on the element and wt.% range, this standard list can be further filtered for elements with large matrix effects using these new controls as seen here:



Of course one can change the conditions and or analytical x-ray lines as necessary. The output is as seen here for standards in the Find list, containing elements with a large fluorescence matrix correction under the current conditions:

FindStandardsFilter: Starting standard filter operation...
FindStandardsFilter: large fluorescence correction for NIST SRM C2402 (Hastelloy C) (651), fe ka (7.3 wt.%), = 0.9176308
FindStandardsFilter: large fluorescence correction for Chromite USNM 117075 (455), ti ka (0.072 wt.%), = 0.9484967
FindStandardsFilter: large fluorescence correction for Chromite USNM 117075 (#455) (856), ti ka (0.072 wt.%), = 0.9485086
FindStandardsFilter: large fluorescence correction for Bornite U.C. #827 (709), fe ka (11.15 wt.%), = 0.9287004
FindStandardsFilter: Standard filter operation complete!

[John Donovan]

New wavescan output format (Goemann-Carpenter), is now available from the Output menu and will load each element/spectrometer to a separate Excel sheet for easy comparison of multiple samples as seen here:

[Paul Carpenter]

John,

In Picturesnap the .img file filtering in the Open file dialog does not show all the .jpg images when you navigate to a new directory.  It seems to require that you go to the all files *.* filtering to work.  It may be that it works when you are in the default run directory but not when you go to retrieve an image from another directory.  It also seems to show .bmp image but not .jpg images when in the .img filter mode.

Cheers,

Paul

[John Donovan]

In Picturesnap the .img file filtering in the Open file dialog does not show all the .jpg images when you navigate to a new directory.  It seems to require that you go to the all files *.* filtering to work.  It may be that it works when you are in the default run directory but not when you go to retrieve an image from another directory.  It also seems to show .bmp image but not .jpg images when in the .img filter mode.

Hmmm, I'm not seeing this issue. I see all the jpg files in my folders as seen here:



So I guess I'm saying I need more information...

Wait, are they .jpeg files that are missing?
john

Edit by John: I should add that although today the .jpg and .jpeg extensions are used somewhat interchangeably, originally the .jpeg indicated Macintosh format.  The difference being that the byte order is different on Mac versus Windows, i.e., big-endian versus little-endian.

To be on the safe side I didn't add .jpeg and .tiff extensions because I'm not sure the VB6 control will always handle these byte flipped images properly, but as you say, you can simply switch to the All Files filter.

[Probeman]

Hi Paul,
OK, I do see this effect, but only in some folders it seems! I do not understand why, but it might be if the default file is displayed. I will investigate.

One can always type *.jpg to see just the jpg files for example.

Edit by John: OK, I see the problem.  If I specify a default file with an extension in the code it only displays files of that type. So I think I can fix that...

[John Donovan]

Hi Paul,
OK, I do see this effect, but only in some folders it seems! I do not understand why, but it might be if the default file is displayed. I will investigate.

One can always type *.jpg to see just the jpg files for example.

Edit by John: OK, I see the problem.  If I specify a default file with an extension in the code it only displays files of that type. So I think I can fix that...

The problem was a subtle aspect of the way Microsoft handles the file filter string for multiple extensions. Latest PFE update will fix the issue. Thanks for reporting this!

[Paul Carpenter]

John,

Thanks man!  You made the change, it had downstream implications and apparently made Kim Jong disappear from public view, who would have known that.  And also a problem with image filtering.

Cheers,

Paul

[Probeman]

On certain Win7 computers (but only on some of them!), when you perform an update from the Help menu, you follow the msi installer prompts to update Probe For EPMA and you will sometimes get this error after clicking the Install button:



After much investigation we have to admit we do not understand this error (we tried all the suggestions we found after googling this error and nothing seems to fix the issue).

However, if you simply browse to the Probe for EPMA application folder (usually C:\Probe Software\Probe for EPMA) and just double-click the ProbeForEPMA.msi or CalcZAF.msi installer that you downloaded, the msi installer  will run fine the 2nd time.

This might be related to recent reports of other Windows installer errors for Skype and Adobe. If you figure out what Microsoft did and know how to fix it, please tell us!

[John Donovan]

For reasons that are not clear to me, the latest update works perfectly on some computers, but on others (with no commonality that I can think of), the update needs to be run twice by clicking the Delete Update button as seen here:



Then just click the Download Update button again.

To make sure that you have the proper update, the version.txt file (accessible from the Help menu also) will have entries dated 12/07/14 or later.

Edit by John: I'm "bumping" the version number from 10.5.4 to 10.5.5 just to insure that the update goes well.  The only feature added is an obscure function that is documented here:

https://probesoftware.com/smf/index.php?topic=40.msg138#msg138

[John Donovan]

The latest version of Probe for EPMA and CalcImage now include new keywords for the ACQ calibration files. These ACQ calibration files contain stage coordinate calibration information for BMP, JPG and GIF image files. Now these .ACQ files will also contain keV and magnification information as well (if they are beam scan images).

The features affected by this change are the PictureSnap | Save As menu, the Run | Display, Annotate and Export Analog Signal Images menu dialog and the Output | Save Images As BMP Files menu.

This change is implemented to facilitate the loading of beam scan images from the CalcImage | Processing | Calibrate Beam Scan window for beam scan calibration. That is, this information will automatically be loaded to the beam scan image keV and mag text fields when the beam scan image is loaded.

The format of the .ACQ file is now as seen in this example:

[stage]
PictureSnap mode= 0
Number of calibration points= 2
Number of Z calibration points= 0
Screen reference point1 (twips)=7680,5760
Screen reference point2 (twips)=0,0
Stage reference point1=16999.29,-21013.83
Stage reference point2=16756.71,-20830.17
X_Polarity= 0
Y_Polarity= 0
Stage_Units=um

[ColumnConditions]
kilovolts= 15
magnification= 1555

For more information on this cool new way to automatically calibrate your instrument magnification for arbitrary kilovolt settings, see this description here:

https://probesoftware.com/smf/index.php?topic=41.msg2074#msg2074

Edit by John:  I also added a scan rotation keyword to the .ACQ files (v. 10.5.7):

[stage]
PictureSnap mode= 0
Number of calibration points= 2
Number of Z calibration points= 0
Screen reference point1 (twips)=7680,7680
Screen reference point2 (twips)=0,0
Stage reference point1=49.9475,50.9015
Stage reference point2=50.7525,50.0985
X_Polarity=-1
Y_Polarity=-1
Stage_Units=mm

[ColumnConditions]
kilovolts= 15
magnification= 400
scanrotation= 9.2


This may be useful as a check when loading beam scan images in the new CalcImage  Processing | Calibrate Beam Scan menu described here:

https://probesoftware.com/smf/index.php?topic=396.msg2074#msg2074