[Sbs_gems] [EXTERNAL] Re: GEM Negative Pulse Overview

[Gnanvo, Kondo (kg6cq)]

Hi Sean, Anu
We discussed at the RC meeting the afternoon that now (swing or owl shift) is the best time to do the studies that you suggested below.

I was wondering if you and /or Anu will be able to come in and perform that study. Any time you will want is ok as.

Sean: On your suggestion below, you wanted to do it with LH2, however we are now taking LD2 data until tomorrow and I assume it is OK to do it with LD2 as well (only to adjust the beam current accordingly) right?

You could sill do 1, 3, 5 and 7 uA on LD2

Let me know
Best regards
Kondo

From: Sbs_gems <sbs_gems-bounces at jlab.org> On Behalf Of Sean Jeffas
Sent: Wednesday, January 26, 2022 4:00 PM
To: Sbs_gems at jlab.org
Subject: [Sbs_gems] [EXTERNAL] Re: GEM Negative Pulse Overview

Hi All,

Here is my ideas for doing a GEM study on this:

For all runs use GM1_Pedestal configuration (full readout, no zero suppression)
200k events for all GEMs (should be 3 minutes or less)
1) 1 uA on LH2 GEMs at operational HV
2) 3 uA on LH2 GEMs at operational HV
3) 3 uA on LH2 GEMs at 1500 V
4) 5 uA on LH2 GEMs at operational HV
5) 8 uA on LH2 GEMs at operational HV
6) 8 uA on LH2 GEMs at 1500 V
7) 8 uA on LH2 GEMs at 0 V

Overall this should take about 30 minutes of beam time. Please let me know if you have any ideas to change or add something to this.

Best,
Sean

On Wed, Jan 26, 2022 at 1:32 PM Sean Jeffas <sj9ry at virginia.edu<mailto:sj9ry at virginia.edu>> wrote:
Hi All,

I am mostly sending this out because Kondo, Andrew, and Bogdan were not in the GEM meeting today and I overviewed the issue of the negative GEM signals seen in the hall. I have attached my slides below.


https://hallaweb.jlab.org/wiki/images/f/fd/Neg_signal_1_26_22.pdf


At the meeting Ben suggested one interesting reason for why the CM would decrease at higher currents, but the fraction of negative pulses would also decrease. He suggested that this may happen if the positive signal "overrides" a negative signal on the same strip. So at higher currents, with higher occupancy, there are less strips with negative signals (because they are already positive), decreasing the fraction, but the negative signals are still used in the CM calculation, causing the decrease in the CM. (Ben, please correct me if I have not described your idea properly).

Overall we need more data and analysis to understand this problem. We should decide on a set of quick runs to study the negative pulses more. This will also be simple because Kondo is RC this week. Kondo has already suggested a test with HV ~1500V and beam on with online zero suppression enabled. I think we should also take some data at different beam currents with nominal HV values and the online zero suppression disabled. This can be accomplished with 30 minutes of beam time. Please suggest any other ideas for tests with beam.


For analysis, there are not many options in the analyzer right now to study the negative signals. We only have a few predefined histograms saved for full readout events, but most of the data is discarded because it is not of interest for the tracking. We will need to add some analysis to store all the strip information to study this properly. Then we can fully see what type of events the negative signals affect on average. We can check if they show up in patterns of strips on all the APVs, their occupancy, and how they correlate to real signals. Please suggest any other analysis ideas for distributions to check.

Best,
Sean
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/sbs_gems/attachments/20220130/a308b74c/attachment-0001.html>