News:

:) Please keep your software updated for best results!

Main Menu

Latest version changes for Probe for EPMA (and CalcImage) v. 13.9.2

Started by John Donovan, August 17, 2013, 12:58:53 PM

Previous topic - Next topic

Philipp Poeml

Quote from: John Donovan on September 26, 2019, 09:51:43 AM
By necessity the Disable Quant flag only applies to the calculation of elements in unknowns, not in standards.

Think of it this way: even if an element is disabled for quantification in an unknown, you'd still want that element utilized in the std k-factor calculation, if it is present in that standard composition.

True.

But it wouldn't be duplicate anymore.
But ok, if it does not affect calculation speed or anything, I guess it does not matter.

John Donovan

Quote from: Philipp Poeml on September 26, 2019, 09:57:01 AM
Quote from: John Donovan on September 26, 2019, 09:51:43 AM
By necessity the Disable Quant flag only applies to the calculation of elements in unknowns, not in standards.

Think of it this way: even if an element is disabled for quantification in an unknown, you'd still want that element utilized in the std k-factor calculation, if it is present in that standard composition.

True.

But it wouldn't be duplicate anymore.
But ok, if it does not affect calculation speed or anything, I guess it does not matter.

The std k-factor code knows nothing about disable quant flags.  But yes, it doesn't affect the speed much so it really doesn't matter.
john

Edit by John: OK, I now notice that the disable quant flags are checked in the code for the std k-factor calculation, but it still has to calculate std k-factor for each (not quant disabled) emitting line, with all the other elements not duplicated and present as specified concentrations.
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

John Donovan

Probeman recently said to us that he wanted to be able to disable/enable individual data points based on their EDS spectra.

It might be because the EDS deadtime was over 90% and the EDS spectra were no longer suitable for quantification, but it also might be that in examining the EDS spectra, it is determined to be necessary to exclude certain data points that are not the phase of interest. For whatever reason.

So we added a disable and an enable button in the Run | Display and Export EDS spectra menu dialog as seen here:



Maybe others will find it useful.
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

Malcolm Roberts

John
I'm having a particular problem that I cannot track easily. I have a win10 machine connected to the new 8530F+ and another for off-line data processing. I am currently having to process the data on line rather than the work station as when I attempt to open the .mdb files acquired on-line with the off line computer I get an error that the database is not is not compatible. It then rolls back and closes. This sounds like some annoying windows type behavior. Anyone seen this already and what should I do about it?? I have 12.7.7 on remote and the previous version on-line. But don't think that this should be an issue after all, I have always opened the .mdb in PfEPMA anyway.
Cheers
Malc

John Donovan

Hi Malcolm,
Is this the problem you are having?

https://smf.probesoftware.com/index.php?topic=105.msg8095#msg8095

Microsoft broke their Access database support last year, but it seems to have been fixed. Is this computer not able to get Windows updates?
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

Malcolm Roberts

Hi John
That's the beast. As for updates, it can but I tend not to let it because these are in the hands of UWA IT "Gods" and we cannot foresake control over things things known to be less about our interests.
Cheers
Malc

John Donovan

Always apply OS updates!

For on-line computers one should select the option to "download updates but let me decide when to install them".  So Windows Update doesn't re-boot your computer in the middle of a run!

For off-line computers just let them install automatically. It's important.
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

Mike Matthews

I've been getting a similar behaviour (but for a different reason I think). When opening a PfE database file over a network connection I sometimes get a 'database already open locked for access' error message. Copying the database file to the desktop solves the problem so I suspect it's something to do with the slightly slower connection over the network causing a false error. Took me a while to work out that the database files weren't getting randomly corrupted.

John Donovan

Hi Mike,
Yeah, that's a different thing.

PFE locks the database when ever it performs a "transaction". That is updating the database.  This is to prevent any possibility of data corruption.

https://en.wikipedia.org/wiki/Transaction_processing

You might notice a .LDB file at times in the folder. That is Microsoft's file locking mechanism, but it doesn't work fast enough over a network connection, so yes, copy the database to a local drive and all is well.
john
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

John Donovan

Recently we improved the code handling the fact that the JEOL EDS interface can support 3 or 4 pulse processing time constants. Older "white type" JEOL EDS spectrometers usually have 3 time constants, while the newer "black type" spectrometers have 4 time constants.

So when PFE first connects to the JEOL EDS it asks how many time constants do you support. The latest PFE version (update any time), re-configures the dropdown list options to reflect this information as seen in the Acquisition Options dialog:



It should also be mentioned that these time constant (throughput) controls shown above may not be active (even you are connected to a Bruker, Thermo or JEOL EDS spectrometer), unless you have the relevant flag set in the Probewin.ini file as shown here in the [hardware] section:

EDSSpectraInterfacePresent=1   ; non-zero EDS spectrum interface feature available (Thermo, Bruker, etc)
EDSSpectraInterfaceType=6   ; 0 = Demo, 1 = Unused, 2 = Bruker, 3 = Oxford, 4 = Unused, 5 = Thermo NSS, 6 = JEOL
EDSMaxEnergyThroughputPresent=1 ; indicate EDS detector get/set max energy and pulse processing

By the way, if you do have a Bruker, Thermo or JEOL EDS detector and would like to integrate WDS and EDS measurements for integrated quantitative analysis please contact us at Probe Software.

For more information on this very cool feature where one can combine acquisition and quantification of elements by WDS and EDS using full spectrum EDS acquisition for maximum flexibility, please see this topic (started in 2013!) and do let us know if you have any questions:

https://smf.probesoftware.com/index.php?topic=79.0

If you would like to jump ahead to a post showing more recent improvements on this integrated WDS and EDS method, specifically demonstrating the ability to perform quantitative spectral interference corrections between WDS and EDS elements, please check this post:

https://smf.probesoftware.com/index.php?topic=79.msg7818#msg7818

Oh, and this latest version fixes a few minor bugs recently reported by Owen Neill and Nick Bulloss. Please update any time using the Help menu in PFE.
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

John Donovan

The latest version of Probe for EPMA has a small tweak that some might find useful.

When plotting your wavescans from the Plot! window, one can model their background fits using the Model backgrounds button. This button displays the Model backgrounds dialog which can be utilized to graphically select various background fits, in addition to specifying the multi-point background positions.

In addition one can assign these selections to the current sample (by default), or to a number of selected samples as seen here:



In this latest version of PFE, not only are the background fits and multi-point background positions saved to the selected samples, but now also the normal hi and lo off-peak background positions.  Previously the hi and lo off-peak background positions were only modified in the current sample.

Of course the software will not modify hi/lo or multi-point background positions for samples which already contain data.
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

John Donovan

Theo Ntaflos, via Gareth Seward, reported to us that on one of his new computers, when he tried to display KLM markers in the multi-point-background display window in Probe for EPMA, he got the following error:



After a bit of research, we discovered this interesting web page:

http://support.sas.com/kb/48/849.html

Apparently on some Windows computers that only have Microsoft Office 64 bit installed, older versions of Microsoft Access databases (MDB files), are no longer supported.  One solution is to load the older Access MDB database file into Microsoft Access and save it as an Access 2000 format database. This may be necessary for Probe for EPMA databases created prior to 2007 or so, as more people install Microsoft Office 64 bit (and Probe for EPMA) on a new computer.

In the case of the XRAY.MDB x-ray database, we have now updated the (NIST) x-ray KLM database for you, to the Access 2000 format in version 12.8.0 of the CalcZAF.msi installer.

So if you see this error from the multi-point background or the Plot! window for displaying wavescans with KLM markers, you will need to update your Probe for EPMA installation by updating your CalcZAF installation. That simply means that you open the CalcZAF application, then click on the Help | Update CalcZAF menu, and download the latest update.
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

John Donovan

As requested by Gareth Seward, we made a small modification to the matrix correction (fluorescence) error trapping code in PFE so that the first time you get an error from a bad point, void, etc., the program will ask if you want to skip these pop-up warnings.

If you click Yes, now it will simply log these errors to the log window as it already does for x-ray map quantification in CalcImage.
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

John Donovan

A very tiny change in Probe for EPMA regarding the "demo" or "simulation" mode concerning 8230 and 8530 JEOL instruments.

A colleague of ours was recently using Probe for EPMA at home in simulation mode to design experimental setups (due to the recent stay at home situation), and was having trouble setting some high beam currents in demo mode.  The issue was that his range of coarse condenser calibration values was too small for setting high beam currents.

Now normally the JEOL simulation (and real time!) mode for 8900/8200/8500/8230/8530 utilizes an iterative method to set the beam current, based on first setting a coarse condenser lens value from a calibration curve (either previously stored or in real time), and then iterating the fine condenser lens value until the desired beam current is reached.

However, starting with version 5 of the EIKS interface, JEOL added a new firmware function to set the beam current internally for 8230 and 8530 instruments using firmware. This is indicated in Probe for EPMA by editing the BeamCurrentType in the Probewin.ini file to be "1". That is:

[hardware]
BeamCurrentType=1

However in simulation mode, PFE was still using the iterative method even when simulating an 8230 or 8530 instrument (and the BeamCurrentType=1). This is now fixed, and so the latest version of Probe for EPMA will now simulate making a firmware call for setting the beam current for 8230/8530 instruments (in demo mode) if the BeamCurrentType=1 in the probewin.ini file.

See here for more details:

https://smf.probesoftware.com/index.php?topic=80.msg7423#msg7423

Hope that is clear.
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"

John Donovan

Ben Wade reported a small bug that occurs when editing standard assignments in a TDI sample with no TDI data. This is now fixed.

Update to the latest Probe for EPMA from the Help menu and you won't get this spurious warning.
John J. Donovan, Pres. 
(541) 343-3400

"Not Absolutely Certain, Yet Reliable"