JEOL Stage Orientation (vs Cameca)

[John Donovan]

As most everyone knows, the Cameca stage is "Cartesian" (from Descartes of course!), because Cameca (a French company of course), uses a "geographical" coordinate orientation system for its stage- that is, maximum X and Y, to the upper right.

Whereas JEOL, uses what I sometimes call an "anti-Cartesian" stage, because its stage coordinate orientation system is exactly the opposite. That is, maximum X and Y, towards the lower left.

Does anyone know if there is any historical reason for JEOL choosing its stage coordinate system?  It's pretty obvious why Cameca choose Cartesian coordinates!

[Dan R]

Hi John-
I'm not sure what the stage looks like for older JEOL instruments, but the coordinate system for the 8530F uses an offset cartesian system (if there is such a thing). There is an origin located at (0,0), but the limits at the upper left are (+45, -60) and the lower right are (-45, +40).
-Dan

[bgarcia]

I can confirm.  Chalk it up to that Japanese "out-of-the-western-european-box" thinking.  They ARE on the other side of the world, after all.

[Probeman]

Yup, we're all saying the same thing!

The difference in the stage between the older JEOL instruments (8900, 8200, 8500) and the newer JEOL (8230, 8530) instruments, is that they moved the origin (0,0) from the lower left on the older instruments, to the middle of the stage for the newer instruments. But the JEOL stage is still "anti-cartesian", (increasing x/y to the lower left)!

If the JEOL instrument was designed in Australia I could understand it being "anti-cartesian".       

Though just maybe Ben has a point and the stage is that way because both Japan and Australia drive on "the other side of the road"?   

[Probeman]

Ok, here's my latest JEOL stage polarity hypothesis:

The Cameca stage (and beam) scans move the stage from left to right (fast scan) and top to bottom just like reading a book (in France).

But in Japan, traditionally, they read Kanji from top to bottom (fast scan) and right to left- just like the stage scans!   

Hence they use anti-cartesian stage coordinates maybe because then you're always scanning "up" the stage axes!  How does that sound?

[Probeman]

At the AMAS 2015 in Hobart (an excellent meeting organized by Karsten Goemann), I had a nice chat with Hideo Takahashi from JEOL about the history of the JEOL EPMA stage.

Turns out that when they first started, Nippon Steel made it a requirement that the upper right corner of the sample was to be designated as 0,0. But because JEOL decided to keep the coordinate system in positive millimeters, this forced the stage orientation to be "anti-Cartesian".

As to why the JEOL stage scans in the Y direction it was also because of Nippon Steel requirements. Apparently early on JEOL made some EPMA instruments for Nippon Steel with a normal x stage width, but in the y direction it was very long to handle large ingots of steel.

[Probeman]

I'm going to call this the CRT/Newspaper theory of EPMA scanning.

When we acquire a beam scan on an EPMA instrument it makes sense that we start in the upper left of the scan area and scan horizontally from left to right stepping down for each scan line until we end in the lower right corner. Why? Because that is the way the first CRT (cathode ray tube) devices were designed to scan.  But why were CRTs designed to scan this way? My thought is this is because when you're printing out characters to the screen (on the first output devices), you want to read it just as one would read a newspaper or book. Start in the upper left, scan to the right, and down one scan line at a time.

As for stage scanning, in the Cameca world the stage is scanned in exactly the same way. From upper left to lower right, just as we do for beam scans. There it should end.

But on JEOL instruments we start scanning in the upper right and scan down, then step horizontally from the right to the left, ending in the lower left. Now maybe JEOL had to scan their stage this way because Nippon Steel requested it as proposed by Takahashi. But why did Nippon Steel request that the stage be scanned this way? Again as previously stated, I propose it's because that's the way a Japanese newspaper in Kanji is read. Starting at the upper right and scanning down.

Now it's true that JEOL instruments use the same scanning convention for beam scans just as Cameca does, but that's only because that's the way CRTs are designed to scan. The first beam instruments could only scan the beam on a CRT. But once stage scanning became an option, the default way to think of doing this is the way we read our respective newspapers. Does anyone have a better theory?

By the way the first beam scanned instrument were developed by Ernst Ruska/Max Knoll/Manfred von Ardenne. See McMullan, D. (1988). "Von Ardenne and the scanning electron microscope". Proc Roy Microsc Soc. 23: 283–288

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

That's my half baked theory of EPMA history for today!