Author Topic: Cyclic patterns in quant data  (Read 220 times)

Ken Orr

  • Student
  • *
  • Posts: 2
Cyclic patterns in quant data
« on: August 19, 2019, 08:25:57 pm »
Hi,

I'm using the polygon extraction tool to extract the quant data from my maps. When I look into the data and start plotting elements up, a cyclic pattern emerges (see attachments). I'm plotting around 40000 points from a single pyroxene grain.
This pattern occurs in all the elements. Does anyone have any ideas as to why this is happening?

Thanks.

Probeman

  • Emeritus
  • *****
  • Posts: 1955
  • Never sleeps...
    • John Donovan
Re: Cyclic patterns in quant data
« Reply #1 on: August 19, 2019, 08:39:13 pm »
Hi Ken,
Just a wild guess here.

Could the raw counts per pixel be close to small integers, e.g., 1, 2, 3, etc. photons?   In other words were these pixels acquired with a very short dwell time per pixel, so only a few photons were captured?  Take a look at the net intensity maps, and look at the count rate...

By the way, what was the pixel dwell time and beam current?
john
« Last Edit: August 19, 2019, 08:41:11 pm by Probeman »
The only stupid question is the one not asked!

Ken Orr

  • Student
  • *
  • Posts: 2
Re: Cyclic patterns in quant data
« Reply #2 on: August 19, 2019, 09:46:41 pm »
The dwell time was 40 ms with 15 kv and 40 amps. And a 2 ┬Ám pixel size. They were 16 hour maps. I don't have net intensity maps as yet. Does 40 ms sound like that might be the issue?

Probeman

  • Emeritus
  • *****
  • Posts: 1955
  • Never sleeps...
    • John Donovan
Re: Cyclic patterns in quant data
« Reply #3 on: August 19, 2019, 10:04:12 pm »
40 millisec is a pretty short time for that beam current (I assume you meant 40 nano-amps!). It could be simply due to counting statistics, once the background is subtracted.

One will see the same effect on an EDS spectrum when zoomed in (see attached).

Take a look at the net intensity maps and let us know what you see.
john
The only stupid question is the one not asked!