Beam map aspect ratio not correct

[Dan R]

I recently installed PI 1.2.3.2016 (and PFE 11.6.1) on our JEOL 8530F. I loaded a previous acquisition file for square beam maps (150x150 pixels, 9500X or ~10x10um). However, now the 'Acquisition' window says the dimensions are going to be 12.54 x 9.06 um -- I ran the maps and the images do come out elongated in the X direction. Is there a way that I can force the pixel size to be square again for beam maps?

[John Donovan]

I recently installed PI 1.2.3.2016 (and PFE 11.6.1) on our JEOL 8530F. I loaded a previous acquisition file for square beam maps (150x150 pixels, 9500X or ~10x10um). However, now the 'Acquisition' window says the dimensions are going to be 12.54 x 9.06 um -- I ran the maps and the images do come out elongated in the X direction. Is there a way that I can force the pixel size to be square again for beam maps?

Hi Dan,
Yes.  You'll need to go into the Setup | Probe Setup menu and enter X and Y values for the beam scan calibration. 

A super accurate method way to get this calibration is by comparing a stage scan with a beam scan image in the new CalcImage Beam Scan Calibration, as shown here:

http://probesoftware.com/smf/index.php?topic=396.msg2481#msg2481

But to begin with, just enter the same values for X and Y so you at least get a roughly square beam scan to start with!
john

PS Note that v. 1.2.4 of Probe Image is now available for download.

[Probeman]

Here's an example from my Sx100 at UofO:



In this example, the values are the same for X and Y, though if I had more carefully calibrated them, they would probably be slightly different.  Note that the scan rotation changes slightly as the mag increases (sorry the fractional decimal point isn't visible in this view).
john

[Dan R]

Thanks John-
I'll modify my files. However I will note that the beam scan itself looks correct based on the sample burn -- I compared this with an identical map using JEOL software.
-Dan

[Probeman]

Thanks John-
I'll modify my files. However I will note that the beam scan itself looks correct based on the sample burn -- I compared this with an identical map using JEOL software.
-Dan

This makes sense.  Because Probe Image bases the aspect ratio calculation (for display) on the beam calibration!  After all, we don't want to *assume* our pixels are perfectly square do we?
john

[Probeman]

Here's an example from my Sx100 at UofO:



In this example, the values are the same for X and Y, though if I had more carefully calibrated them, they would probably be slightly different.  Note that the scan rotation changes slightly as the mag increases (sorry the fractional decimal point isn't visible in this view).
john

Bringing this post back for someone.  I should point out that the above example on my Sx100 shows a square image because the number of X and Y calibration (microns) are the same. But because Probe Image utilizes square pixels (unlike Probe for EPMA which has a separate aspect ratio parameter), to obtain a say 4:3 image aspect ratio, one simply determines the X and Y FOVs (in microns) and enters them in the above Beam Scan Calibration dialog in Probe Image from the Setup | Probe Image Setup menu.

One can easily calibrate their beam scan using the new beam scan calibration menu in CalcImage here:

http://probesoftware.com/smf/index.php?topic=396.0

by simply acquiring both a stage map and a beam scan on roughly the same area (where one can see three features on both images) and calibrating it using three fiducials as described in the link above.  In CalcImage this is the Processing | Calibrate Beam Scan From Stage Scan menu.