Wish List for PFE Features

[Philipp Poeml]

Hi John,

it seems that one cannot use "quick standards" with "digitized sample setups".

To use the combination would be super useful. Could you implement that?

Thanks
Philipp

[Probeman]

Hi John,
It seems that one cannot use "quick standards" with "digitized sample setups".

To use the combination would be super useful. Could you implement that?

I assume you are getting this message?



Actually it gets very complicated... can you tell me what exactly you are trying to change in the standards by using sample setups? If it's just the analytical conditions, as the message says, you can use "quick standards" with the Automate! Digitized Conditions acquisition option.

Also be aware that "quick standards" have some nice options themselves in the Acquire! | Acquisition Options dialog as seen here:

http://probesoftware.com/smf/index.php?topic=8.msg1095#msg1095

[Anette von der Handt]

1) Would it be possible to have another function where I can first include/exclude elements and then get a list of the filtered standards with their Z? Seems to me just like a combination of what is already available in the program..

2) In standard.exe, I can check for nominal interferences in a given standard composition. However, I cannot choose to specify an element which is not in the element list of the standard. It would be great if I could just choose random elements to quickly verify that I do not have a peak interference in my MAN standard.

I understand #2 well enough. Remember though, the nominal interference calculation assumes Gaussian overlaps and so is not very accurate.

The best practice in my mind is to simply acquire all analyzed elements in all standards (the default non quick standards), and then analyze them (as if they are unknowns from the Analyze! window) and see if anything shows up that shouldn't be there. 

Actually even easier is to simply unselect any standard intensities that plot above the general MAN curve trend. After all, background is *by definition* the lowest intensity one can measure, so if the std intensity point is above the trend, it's either a contamination or an interference and you really don't care which it is- just deselect it from the fit.

This certainly works well but in terms of efficiency (time=money) I rather would like to not spend time acquiring data I ultimately will not use. Just having access in this dialogue to the list of nominal interferences would serve as a nice guideline to weed out the obvious non-candidates.

As for #1, maybe you should provide an example of what you mean.

Let's say some user wants to map some multi-phase material with a wider spread of mean Z for various major and minor elements, in total 10 elements. And our metal standard block is in a somewhat rough shape ,  the user does not know our various standards by heart and I, microprobe mastress, am not around. And time is money again (see above).

Right now standard.exe allows me (1) to look up standards that contain a certain element (Find..) or (2) to list all standard names and average Z and. I just thought having the option to combine (1) and (2) in one output or - for the purpose of MAN backgrounds - exclude (rather than find) standards containing a certain element(s) and then list their name and Z.

However, I am aware that I could probably create an excel spreadsheet that would contain this kind of information, so overall this function would be a "nice to have".

Thanks!

[Philipp Poeml]

2) In standard.exe, I can check for nominal interferences in a given standard composition. However, I cannot choose to specify an element which is not in the element list of the standard. It would be great if I could just choose random elements to quickly verify that I do not have a peak interference in my MAN standard.

I understand #2 well enough. Remember though, the nominal interference calculation assumes Gaussian overlaps and so is not very accurate.

The best practice in my mind is to simply acquire all analyzed elements in all standards (the default non quick standards), and then analyze them (as if they are unknowns from the Analyze! window) and see if anything shows up that shouldn't be there. 

Actually even easier is to simply unselect any standard intensities that plot above the general MAN curve trend. After all, background is *by definition* the lowest intensity one can measure, so if the std intensity point is above the trend, it's either a contamination or an interference and you really don't care which it is- just deselect it from the fit.

This certainly works well but in terms of efficiency (time=money) I rather would like to not spend time acquiring data I ultimately will not use. Just having access in this dialogue to the list of nominal interferences would serve as a nice guideline to weed out the obvious non-candidates.

I just use in standard.exe the "enter unknown" window for this purpose. It allows me to type in any composition I want and to do a thorough intereference check on it for peak and background postitions.

[Philipp Poeml]

Hi John,
It seems that one cannot use "quick standards" with "digitized sample setups".

To use the combination would be super useful. Could you implement that?

I assume you are getting this message?



Actually it gets very complicated... can you tell me what exactly you are trying to change in the standards by using sample setups? If it's just the analytical conditions, as the message says, you can use "quick standards" with the Automate! Digitized Conditions acquisition option.

Also be aware that "quick standards" have some nice options themselves in the Acquire! | Acquisition Options dialog as seen here:

http://probesoftware.com/smf/index.php?topic=8.msg1095#msg1095

Yes, that is the message I get. I am trying to measure standards and unknowns in one run and also remeasure the standards after a certain time. For this I need to:
*change the beam current
* change the counting times of some elements
* change the beam size
* set some elements to "acquisition-disabled" (and later re-enable them)

I figured the best way to do this is using "digitized sample setups".
Now, without quick standards, I waste A LOT of time for analyzing all elements over and over again...

[Philipp Poeml]

Hi John,

Why not display spectro units here (screenshot)? You do that for the off peak interferences. Nice would be also spectro units relative and absolute... Also in calczaf.exe where I do these calculations from time to time.

Thanks!



Edit by John: Your request has been implemented in v. 10.5.4

Great, John, thanks a lot. Now, the next wish:
A little tick box: "Tick this box to see the interefence lines in realtive spectro position offsets from the peak".
Would be SUPER handy!

[John Donovan]

Yes, that is the message I get. I am trying to measure standards and unknowns in one run and also remeasure the standards after a certain time. For this I need to:
*change the beam current
* change the counting times of some elements
* change the beam size
* set some elements to "acquisition-disabled" (and later re-enable them)

I figured the best way to do this is using "digitized sample setups".
Now, without quick standards, I waste A LOT of time for analyzing all elements over and over again...

Here's what I would do:

1. Set up the counting times on the standards and unknowns differently using the unknown count factors in the Count Times dialog, see here:

http://probesoftware.com/smf/index.php?topic=29.msg397#msg397

2. Using the Digitized Conditions dialog in the Automate! window, specify one set of beam currents and beam sizes for your standard positions and different conditions for your unknown positions.

3. Select "Use Quick Standards" in Automate!

4. As for "Acquisition disabled" flags- go ahead and set them, and then unset them later!

That is all you have to do for your list above!
john

[Philipp Poeml]

Yes, that is the message I get. I am trying to measure standards and unknowns in one run and also remeasure the standards after a certain time. For this I need to:
*change the beam current
* change the counting times of some elements
* change the beam size
* set some elements to "acquisition-disabled" (and later re-enable them)

I figured the best way to do this is using "digitized sample setups".
Now, without quick standards, I waste A LOT of time for analyzing all elements over and over again...

Here's what I would do:

1. Set up the counting times on the standards and unknowns differently using the unknown count factors in the Count Times dialog, see here:

http://probesoftware.com/smf/index.php?topic=29.msg397#msg397

2. Using the Digitized Conditions dialog in the Automate! window, specify one set of beam currents and beam sizes for your standard positions and different conditions for your unknown positions.

3. Select "Use Quick Standards" in Automate!

4. As for "Acquisition disabled" flags- go ahead and set them, and then unset them later!

That is all you have to do for your list above!
john

Hi John,

this unknown count time factor is a really cool thing I did not know about. I will check it out. Thanks.

However, I don't see how I can enable/disable elements using your proposal. I am measuring some standards for some virtual standards, and I don't want to waste time measuring these standards of elements that are not present in my sample on the unknown.

I really would like to use the sample setups with "Quick Standards"!! Would solve a lot of problems.

Edit by John: I can understand, but the code is very complicated in this regard and I'm just one guy! Let me think about it and maybe I'll come up with something cool and easy!

[John Donovan]

Great, John, thanks a lot. Now, the next wish:
A little tick box: "Tick this box to see the interefence lines in realtive spectro position offsets from the peak".
Would be SUPER handy!

Your wish is granted (no need for a checkbox)!

[Ben Buse]

Nth background in acquistion options. I regularly use it for standards just measuring the background on the first point. I wish the default when selecting for individual elements was to Yes for Nth Background so that you don't need to turn it on for each element. In which case it would be just controlled by the Nth background on/off overide check box. Maybe an idea for the furture would be to have an option to use Nth backgrounds for standards but not unknowns - so that you can't forgot to turn it off when you go to your unknown samples.

[John Donovan]

Nth background in acquistion options. I regularly use it for standards just measuring the background on the first point. I wish the default when selecting for individual elements was to Yes for Nth Background so that you don't need to turn it on for each element. In which case it would be just controlled by the Nth background on/off overide check box. Maybe an idea for the furture would be to have an option to use Nth backgrounds for standards but not unknowns - so that you can't forgot to turn it off when you go to your unknown samples.

Hi Ben,
I have to be careful here because some people depend on the current paradigm.

However, maybe I could implement a method whereby if you click the Use Nth Point For Off-Peaks checkbox and if all the channels are not using Nth point backgrounds already, I could set them all as Nth point backgrounds.

Likewise, if you uncheck the  Use Nth Point For Off-Peaks checkbox, I could set all channels to not use Nth point backgrounds, if they are all already set to use Nth point backgrounds.

Would that work for you?
john

[Ben Buse]

Hi John,

Yes I'm not sure what the best way is. Or what most people do. For as you point out you could imagine a situation where you want to toggle on or off using the check box the 5 of 10 elements which your using nth background for.

I suppose would it be possible to configure the window so you can select multiple elements at once and change the nth yes/no for the multiple selected elements. Like the really nice feature in setting count times where you can select multiple elements by hold and drag and change the count times for all selected elements

Thanks

Ben

[John Donovan]

Yes I'm not sure what the best way is. Or what most people do. For as you point out you could imagine a situation where you want to toggle on or off using the check box the 5 of 10 elements which your using nth background for.

I suppose would it be possible to configure the window so you can select multiple elements at once and change the nth yes/no for the multiple selected elements. Like the really nice feature in setting count times where you can select multiple elements by hold and drag and change the count times for all selected elements

Hi Ben,
I modified the Nth point select code in the Acquisition Options.  This will be in v. 10.7.4 of PFE.  See if this behavior is more to your liking.

[Ben Buse]

Hi John,

This new behaviour is great. If all the elements are the same it toggles them all off or on. Whereas it elements are doing different things (only some using Nth background) it just disables or enables.

I didn't get originally meant

Thanks

Ben

[John Donovan]

Hi John,

Another nit-picky thing to bug you about...

I have been trying to change the off-peak correction type (linear, polynomial, exponential, etc.) from Analyze! --> Elements. Usually this is easy, but I have sporadically been encountering an error related to the Multi-Point Background positions being outside the spectrometer range. Screenshot is attached - the L-values and positions #'s change, but the overall message is the same.

I can't check what the MPB positions are directly, as these samples were not acquired with MPB's and those options are inactive through Analyze --> Elements. But I did "clone" the sample setup through Acquire, and several of the MPB positions are indeed outside the spectro limit. These must be the defaults, as I did not change, click on or use any of the MPB options during setup and acquisition.

Assuming the out-of-range MPB positions is something I've unwittingly screwed up, is there any way the Elements window could ignore the MPB conditions when if you're trying to switch from, say, linear to polynomial backgrounds? Or ignore checking the MPB parameters when post-processing non-MPB samples?

Thanks!

Hi Owen,
The zero in the error message indicates that this is a zero spectro position, so that is why it's flagging it as out of bounds.  Note sure why there is a zero position there but I can add code to trap for that and load a default.
john