EPMA - Method Development Tool

[Sandrin Feig]

The "EPMA - Method Development Tool" is a database of maximum range wavelength scans of more than 170 of the most common standard materials. It was created to support lab managers and users of electron microprobe facilities with the setup of analyses programs as well as for teaching purposes. The scans were collected on PC2, PC1, PC0, TAP, PET, LPET and LLIF diffracting crystals using a CAMECA SX-100 electron microprobe at the Central Science Laboratory, University of Tasmania, Australia.

The tool can be accessed using the following link: epma-mdt.csl.utas.edu.au

For any comments, suggestions or feedback please contact Sandrin Feig (epma.mdt@utas.edu.au) or post it here



The tool is particularly useful in identifying interferences and to determine suitable background positions.



And also shows edges and holes quite well.

[Probeman]

Very cool!

Now you just need some wavescans from JEOL instruments...  can you provide us with details on what you need format-wise to incorporate JEOL wavescans?

[Sandrin Feig]

Yes, it would be good to also add JEOL wavescans to the page and I am planning to do that. There are a few options I can think of at this stage:

1: During the EPMA2016 conference somebody suggested to add a second x-axis with JEOL units. Although I am not entirely sure to what extend recalculated CAMECA scans can be applied to JEOL instruments, I will definitely give it a go and add the axis. Maybe once it is implemented, somebody with a JEOL instrument can give some feedback on that.

2: JEOL users send me their wavescans and I include them into the tool. For that I probably have to create a second database and we have to define certain conditions to make the scans comparable and compatible (range, step size, peak position to avoid offsets, diffracting crystal range to be covered, etc.). At the moment I am using csv files and I can only load one file at a time. To be able to compare scans, the scans have to be in the same file and use the same x-axis values!

3: I take my standards to a JEOL instrument and run them all. I have already quite a few updates for this page in mind, but after that I will try to get access to a JEOL and apply for money to do the scans. I kind of like this idea since it means that all scans were acquired using the same standards and the same conditions. However, it will take some time till I can collect the scans.

[Probeman]

Hi Sandrin,
I don't think it's necessary to display both JEOL and Cameca wavescans at the same time.

What I would do (if it were my app), would be to have two options: JEOL or Cameca. And depending on which option was selected, the app would only list the wavescans from the appropriate database.  And when the user select a different instrument option the app would clear the display and load the available wavescans for that instrument type.

Of course another approach would be to have a third mode which displays wavescans from both instruments, and for that I would simply convert the x-axis data for both instruments to angstroms (or keV) and then one could display them together.

Question: is the csv format you are using the output from Probe for EPMA?  If so, that would make it easy for other users to provide you with more wavescans for import.  Perhaps I could add a txt file to the wavescan output which could contain the condition parameters for each wavescan...?
john

[Sandrin Feig]

Hi John

Yes, that is exactly what I have in mind. 
Once we have a comprehensive database of JEOL wavescans, there is no need to mix them with CAMECA scans anymore and they can definitely be in separate files, where the user can than choose between JEOL and CAMECA. Especially if I end up capturing scans on the same standards that currently exist for CAMECA my database.

But the problem is that at the moment all scans that you want to display have to be in the same csv file using the 1st column for the x values and the other columns for the individual scans. Furthermore, all wavescans have to be collected with the same x values since I only have one column for them. If I find time, I will look into transferring the csv files over to sql.

The csv format that I am using is fairly similar to the Probe for EPMA output file. In my csv files the first row contains the labels (e.g., standard or mineral names). The first column has all the x values and the other columns carry the wavescans. When I export wavescans collected with your software I basically just have to delete the multiple x columns and adjust the label a little bit, because that is what comes up in the legend under the graph.

At the moment, to be able to easily add wavescans from other labs, the scans should be from 23000 to 80000 with # Points 2048 in the acquisition window (and I use a time of 0.5 sec.). Using these parameters, our instrument than produces scans that start at 23000 and run with a step size of 27 units (23000, 23027, 23054, etc.) to ~78242. Although the end is not too important. The important bit is the step size so that I can use the same x values for everything.

[Probeman]

Hi Sandrin,
Too bad you made the format so "rigid". It seems to me that such an app should be able to handle multiple scans with different numbers of points and x-axis values... and couldn't you recalculate the x-axis values (spectrometer position) "on the fly" based on the instrument type?

By the way, if you are interested I could "ask donovan" to implement a special output option in PFE for exporting exactly the wavescan output one needs for your app.   I assume you'll want (in addition to the sample label), fields for instrument type, keV, beam current, etc., etc.

Wouldn't that be cool?
john

[Sandrin Feig]

Hi John

Well, if you could have a chat with “Donovan” about a slightly different output format that would be awesome!

The format is partially controlled by the csv file and the fact that I can only load one at a time and by the graph gadget that I am using. I will look into changing the format to be a bit more flexible with the x values.
When I started programming the website my focus was more on getting it to work in the first place. And I had to get it ready for the conference…
I will definitely add it to the long list of things that will be optimized soon.

I had a brilliant idea this morning. Should I add some “mystery” scans? These scans could be used for teaching/testing students, What do you think?
Cheers
Sandrin

[Probeman]

Hi John

Well, if you could have a chat with “Donovan” about a slightly different output format that would be awesome!

The format is partially controlled by the csv file and the fact that I can only load one at a time and by the graph gadget that I am using. I will look into changing the format to be a bit more flexible with the x values.
When I started programming the website my focus was more on getting it to work in the first place. And I had to get it ready for the conference…
I will definitely add it to the long list of things that will be optimized soon.

I had a brilliant idea this morning. Should I add some “mystery” scans? These scans could be used for teaching/testing students, What do you think?
Cheers
Sandrin

Hi Sandrin,
I think "Donovan" knows something about "the long list of things" as well!   

Some mystery scans would be useful for teaching I think.

Please provide "Donovan" with a formal criteria of what you would like for an import file.  This doesn't have to be any time soon- you should think through future ideas for implementation now that you have v. 1 up and running. One suggestion for the next import file format: add a version number to the import file, so it can be added to or modified later while maintaining compatibility with previous import file formats.  One easy way to do this would be to just add a version number to the first line and new stuff at the end of the file.

Again, great work.  I think this will be useful.
john

[Sandrin Feig]

Version 1.1 is now online

Most important new feature is the unit selection option for the x-axis. You can now enjoy the scans also with JEOL units (mm), or KeV, or angstrom...


Cheers
Sandrin

[John Donovan]

Version 1.1 is now online

Most important new feature is the unit selection option for the x-axis. You can now enjoy the scans also with JEOL units (mm), or KeV, or angstrom...


Cheers
Sandrin

Hi Sandrin,
Very nice!

So have you got a well defined import format you can provide for us?  I'm wondering if I should implement a method in Probe for EPMA for wavescan output that will import easily into your web page...  what do you currently do in PFE to create an import file for your tool?
john

[Mike Matthews]

Is there a way of running this tool off-line? Unfortunately I'm one of the few people who doesn't have the luxury of an internet connection at my instrument.

P.S. The country drop-down list doesn't seem to work properly on a Mac - it doesn't give any options to select so the registration can't be completed. Worked OK on my pc laptop though.

[Sandrin Feig]

Hi Mike

Unfortunately, at the moment there is no way of running it offline, sorry.

Thanks for pointing out the problem with the Mac and the country list. May I ask which browser you are using. It might be more a browser rather than a Mac problem.

I remember your presentation in Sydney at the conference a couple of years ago. If you send me wavescans of the rather uncommon material you are working with I can include them into the tool. I guess that would make the tool a bit more useful for you. Depending of course if you are allowed to share them.

Cheers
Sandrin

[Sandrin Feig]

Version 1.1 is now online

Most important new feature is the unit selection option for the x-axis. You can now enjoy the scans also with JEOL units (mm), or KeV, or angstrom...


Cheers
Sandrin

Hi Sandrin,
Very nice!

So have you got a well defined import format you can provide for us?  I'm wondering if I should implement a method in Probe for EPMA for wavescan output that will import easily into your web page...  what do you currently do in PFE to create an import file for your tool?
john

Hi John

The format of the CSV-file that I import into the tool is very simple at this stage.



The first column has the x-values and any consecutive column has y-values of the wavescans.  The first row contains the titles of the scans that will appear in the legend of the graph. That is all. However, I am wondering, if we should export wavescan conditions as well (in a separate file?). That might be useful some time down the track...

If Donovan could add such an export function to his software, that would be awesome! I will try to find a way to allow users of the tool to temporarily upload the scan of the material they are working with for comparison with the scans in the tool. I think that would add another interesting option.

Cheers
Sandrin

[Sandrin Feig]

Hi Mike

Unfortunately, at the moment there is no way of running it offline, sorry.

Thanks for pointing out the problem with the Mac and the country list. May I ask which browser you are using. It might be more a browser rather than a Mac problem.

I remember your presentation in Sydney at the conference a couple of years ago. If you send me wavescans of the rather uncommon material you are working with I can include them into the tool. I guess that would make the tool a bit more useful for you. Depending of course if you are allowed to share them.

Cheers
Sandrin

Hi Mike

I think I known what the problem is, you select Europe as a region, but UK is not part of Europe anymore...

I managed to reproduce the error on a Mac with Safari. It works fine with Chrome.
I will modify the code so that it also works with Safari.
Thanks again.
Cheers
Sandrin

[John Donovan]

I think I known what the problem is, you select Europe as a region, but UK is not part of Europe anymore...

I managed to reproduce the error on a Mac with Safari. It works fine with Chrome.

I will modify the code so that it also works with Safari.

Hi Sandrin,
Welcome to the world of pain- sorry, I mean code development!
john