CPM --> μSv and Log10 scale on Map?

36 views
Skip to first unread message

Michael Hittle

unread,
Jan 20, 2016, 1:26:55 AM1/20/16
to Safecast Device Discussions and Support

I can't seem to find the conversion used to determine the bGeigie readings, aside from one reference in 2011 that states a factor of 350. This seems consistent with the OLED readings, but it's unclear if the conversion is as simple this.  Does anyone know precisely what conversion is used?

In addition, the SafeCast map lists a log 10 scale in the color key, yet it's ambiguous as to whether this applies to the entire range, or just the range between 10.09 and 65.54 μSv.  Mapped bGeigie readings seem to vary from other readings at Chernobyl, and I'm wondering if a conversion is required to compare the two. 

Is there anyone that can provide clarity on this? 

Kalin KOZHUHAROV (Safecast)

unread,
Jan 20, 2016, 5:35:34 AM1/20/16
to Michael Hittle, Safecast Device Discussions and Support
On Wed, Jan 20, 2016 at 3:26 PM, Michael Hittle <mhi...@gmail.com> wrote:
> I can't seem to find the conversion used to determine the bGeigie readings,
> aside from one reference in 2011 that states a factor of 350. This seems
> consistent with the OLED readings, but it's unclear if the conversion is as
> simple this. Does anyone know precisely what conversion is used?
>
The official Safecast conversion factor DRE2CPM=334.0 as
(unfortunately) hard-coded in various places in the code (you divide
the cpm to 334 to get DRE in uSv/h).

> In addition, the SafeCast map lists a log 10 scale in the color key, yet
> it's ambiguous as to whether this applies to the entire range, or just the
> range between 10.09 and 65.54 μSv. Mapped bGeigie readings seem to vary
> from other readings at Chernobyl, and I'm wondering if a conversion is
> required to compare the two.
>
While it looks like log10, it is not.

It uses a lot more complex LUT, I almost had blood out of my ears when
trying to reimplement it (joking). For a "simple" way to plot it, have
a look at this code:
https://github.com/thinrope/fixed_sensor_visualization/blob/master/LUT.gpl
It tries to preserve detail at the low end of the values as well as
still provide differences at the high end.

> Is there anyone that can provide clarity on this?
Have you looked at
https://github.com/Safecast/Tilemap/blob/master/README.md for a
general overview (and the linked documents)? That and the code that
actually produces the maps is mostly in that repository.

Regards,
Kalin.
Reply all
Reply to author
Forward
0 new messages