[Color_transp] [EXTERNAL] BCM4A weighted average

[Dipangkar Dutta]

Hi John,
Are you using the number of  reads or the actual increment In the scaler
value itself.
It would be more correct to use the increment in the scaler reading.
For example say during a beam trip there was 3 scaler read and the end of
the 3rd read the reading of the scaler 1000 then the beam turns back on and
it stayed on for 3 reads, the reading at the 6th read is 99000
Then the weight for those 3 scaler reads when the beam was on should be
1/(99000-1000)
And so on for the rest of the beam on reads.
Cheers
Dipangkar


On Tue, Feb 4, 2020 at 5:14 PM John Matter <jcm6fv at virginia.edu> wrote:

> Thank you for taking a look.
>
> For the weights, I was initially using each region's uncertainty but I was
> calculating the uncertainty wrong. I fixed this, but these runs still
> looked off.
>
> I changed the weighting to be the number of scaler reads in the region
> when the beam is not tripped. I believe this is what we discussed in a
> recent meeting.
>
> I updated the elog entry with the new values.
>
> https://hallcweb.jlab.org/elogs/Color+Transparency/33
>
> - John
>
>
> On Mon, Feb 3, 2020 at 7:02 PM Dipangkar Dutta <ddutta07 at gmail.com> wrote:
>
>> Hi John,
>> I looked through your plots. Most of them make sense. But there are a
>> few runs which seem to be lower than that one expects from visual
>> inspection, sometime they are quite a bit lower.
>> And the pattern seems to be that those runs where the blue point is
>> larger than the red point they seem to have a problem (ie the av.
>> weighted current seems too low).
>> Specifically, look at runs 2279, 2283, 2429, 2430, 2431, 2432, 3188
>> Cheers
>> Dipangkar
>>
>> On Mon, Feb 3, 2020 at 3:14 PM John Matter <jcm6fv at virginia.edu> wrote:
>> >
>> > Hi all,
>> >
>> > I just posted some slides about the BCM averaging. If the averages look
>> sensible to everyone, I will update the spreadsheet.
>> >
>> > https://hallcweb.jlab.org/elogs/Color+Transparency/33
>> >
>> > - John
>> > _______________________________________________
>> > Color_transp mailing list
>> > Color_transp at jlab.org
>> > https://mailman.jlab.org/mailman/listinfo/color_transp
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/color_transp/attachments/20200204/5063ca41/attachment-0001.html>