Peaking

[Dan MacDonald]

Hi everyone:

I am a beginner with this software, and am getting hung-up on peaking.  Specifically, when I try to peak  the K Ka line using a PETJ crystal (I have a JEOL 8200), the PFE software tells me that, "Spectro 1 Fine Scan FAILED - Calculated centroid (120.478) is too close to high range (120..572)" and gives me "the same start and Stop values for the K Ka peak at 199.776.  Is there any way I can force the software to accept the correct Peak position at 120.478 (confirmed peak position be doing a peak search in the JEOL software) or to change the default Start Peak  (and high range alue) value in the software to what the spectrometer is actually finding?  Thanks very much for all your help.

Best regards

Dan

[John Donovan]

I am a beginner with this software, and am getting hung-up on peaking.  Specifically, when I try to peak  the K Ka line using a PETJ crystal (I have a JEOL 8200), the PFE software tells me that, "Spectro 1 Fine Scan FAILED - Calculated centroid (120.478) is too close to high range (120..572)" and gives me "the same start and Stop values for the K Ka peak at 199.776.  Is there any way I can force the software to accept the correct Peak position at 120.478 (confirmed peak position be doing a peak search in the JEOL software) or to change the default Start Peak  (and high range alue) value in the software to what the spectrometer is actually finding?  Thanks very much for all your help.

Hi Dan,
We have to get you some training!

Does the PETJ spectrometer have a limited extent?  What is the normal range of a PETJ spectrometer?   It seems to be complaining that your found peak position is too close to the high side of the scan, but then it should automatically try and run a coarse scan with a wider range.  Is it trying a coarse scan after the first warning?

By the way, this error is not in my software "the same start and Stop values for the K Ka peak at 199.776".  Also the K Ka peak on PETJ should be around 119...

If you go into the Peak Scan Options dialog and click on K and click OK, does it give you any errors or warnings?  While you are in this dialog, try changing the "Peaking/ROM Start Size" to a larger number.  If it's 0.1 mm, try setting it to 0.2 mm or so. 

That should produce a twice as wide scan a starting fine scan, but again, it should automatically try a 2nd coarse scan if it fails on the first fine scan so not sure why it won't.
john

[John Donovan]

Also, while you're in the Peak Scan Options dialog, note the yellow button named "Set On-Peak To Current".

Clicking this button will set the on-peak to the current spectrometer position.... but in any case the peaking in PFE should work fine also so I'd like to track this down with your help.

Can you send me this MDB file to my probe software email?
john

[Dan MacDonald]

Good afternoon, John:
You are indeed correct that I need training - I have been asking for it since I started here, but they have not followed through with it.  I am supposed to be taking the next 2 weeks or so to figure it all out ( a bit over optimistic, perhaps).

The normal range of the spectrometer with  the PETJ crystal is 68 to 259 mm (L-value), and it has no other limit restrictions on it.  I think it did try to do a second scan after the first message came up.

I apologize for the poor communication on my part - the software didn't print out an error regarding the Start and Stop peaks, rather it printed out this message (copied and pasted from the main screen):

ROM Brent's Maxima Peak Center Results (cps):
 Element  Spectr  Peaked   StartPk    StopPk  Offset    StartI     StopI StartPB  StopPB
    k ka   1 PETJ   Yes     119.776   119.776 

In most spectroemters, you would expect K Ka to be around 119, but on our probe, that channel has always given us trouble, so we've had to operate with a work-around (it's a long, sordid saga).

When I click on K Ka in the Peak/Scan options window, no warnings come up, and the Peaking/ROM Start Size is 0.152075 - I will try with a larger value, as you suggest.

I also tried under Peaking Options checking the "Display Spectrometer Pre-Scan for Confirmation: and then scrolling over, closer to the 'desired' value for K Ka and then proceeded with the peaking, and it did a better job of finding the peak.  I will try your other suggestions, though, and I will e-mail the MDB file to you - it is just a copied file that was originally entitled the pyroxene.mdb.

Thanks once again for all your kind help - I really appreciate it.

Best regards

Dan

[John Donovan]

I apologize for the poor communication on my part - the software didn't print out an error regarding the Start and Stop peaks, rather it printed out this message (copied and pasted from the main screen):

ROM Brent's Maxima Peak Center Results (cps):
 Element  Spectr  Peaked   StartPk    StopPk  Offset    StartI     StopI StartPB  StopPB
    k ka   1 PETJ   Yes     119.776   119.776 

Hi Dan,
OK, just glad that it didn't think the K Ka peak on PET was 199!   The start and stop peak positions will be the same if the peaking failed or was canceled, so that is normal.

When I click on K Ka in the Peak/Scan options window, no warnings come up, and the Peaking/ROM Start Size is 0.152075 - I will try with a larger value, as you suggest.

The warnings would occur when you click OK, not when the dialog opens...

I also tried under Peaking Options checking the "Display Spectrometer Pre-Scan for Confirmation: and then scrolling over, closer to the 'desired' value for K Ka and then proceeded with the peaking, and it did a better job of finding the peak.  I will try your other suggestions, though, and I will e-mail the MDB file to you - it is just a copied file that was originally entitled the pyroxene.mdb.

You shouldn't need to do a pre-scan.  120 is close enough to 119.775 that it should find the peak on the 2nd scan.  Owen will correct me if I am wrong on this distance.

Once you do find the K Ka peak you can re-use that values by using the Load File Setup button under New Sample Setup button in the Acquire! window.

I don't know where the pyroxene.mdb file came from, but you might want to start a new MDB from scratch. There are some excellent tutorials here that you might want to run through:

http://probesoftware.com/smf/index.php?topic=106.0

It says for quant mapping, but the PFE procedures are exactly the same whether it's point analyses or x-rays maps.
john

[John Donovan]

Hi Dan,
I looked at the MDB file and don't see anything amiss.  I suspect that if you specify a larger start size for that spectrometer the peaking will work on the first scan. If not it should do a coarse scan automatically to find the peak and then a final 2nd fine scan.
john

[John Donovan]

120 is close enough to 119.775 that it should find the peak on the 2nd scan.  Owen will correct me if I am wrong on this distance.

Starting with a K Ka peak position of 120, and using a start size of 0.152075, the scan range should be ~119.25-120.65. So if the pre-peaking position 120, and the actual peak is at 119.775, and a start size of 0.152075 is used, it should find the actual peak without needing the coarse scan.

Hi Owen,
Thanks for chiming in.  I suspect that the peak is at 120.478 based on this warning Dan posted in his first post: "Spectro 1 Fine Scan FAILED - Calculated centroid (120.478) is too close to high range (120.572)"

So, yes I agree with you, that if he uses a start peaking size of 0.18 to 0.22 or so, PFE should find the peak in the first scan attempt.  But if it doesn't, as long as he allows the coarse scan to run, it should still find the peak center eventually.
john

[Dan MacDonald]

I'd like to thank you all for so much useful help.  I have worked through your various solutions and am happy to report that the problem seems to be solved - most of the problem has to do with my lack of experience with the software.

Best regards

Dan