[Sbs_gems] [EXTERNAL] Mixing System for GEMs

[Jack Segal]

Kondo,
   I think the original attempt to use the mixing system is a little more complicated. The plumbing on the buffer
tank is less than ideal. I had put it together with the view to how it would operate when things are running smoothly.
I had not really thought about how it would handle something like recovering from the Argon running out, or something
like that. Both the in and out of the buffer tank are on the top of the tank. So, if the mix in the buffer tank is
starting out in some bad state it takes a long time for the mix in the buffer tank to reach the correct mix desired.
About a day and a half to two days.
   My guess is we started with a bad mixture in the buffer tank and did not have enough time to clear it out.
   I added a purge valve on the exit of the buffer tank. So, we can valve off the delivery to the hall and force four
or five cycles through the buffer tank to purge it. That will take about 20 minutes, or so.

On 11/8/2021 12:43 PM, Gnanvo, Kondo (kg6cq) wrote:
> Hi Zeke & Jack,
> Thanks for performing the tests and summarizing the results.
> 
> Zeke:
> I am a bit confused by your statement, maybe I did not understand very well
> 
> Is it your conclusion that the efficiency drop that we saw in the GEMs a few weeks ago is explained by the 2% disconnect between the mixture ratio readings and the actual ratio?
> Best regards
> Kondo
> ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> *From:* Sbs_gems <sbs_gems-bounces at jlab.org> on behalf of Ezekiel Wertz <ewertz at email.wm.edu>
> *Sent:* Monday, November 8, 2021 12:16 PM
> *To:* Bogdan Wojtsekhowski <bogdanw at jlab.org>
> *Cc:* sbs_gems at jlab.org <sbs_gems at jlab.org>; Jack Segal <segal at jlab.org>
> *Subject:* Re: [Sbs_gems] [EXTERNAL] Mixing System for GEMs
> Hi Bogdan,
> 
> So what it seems like to me is that the different techniques used in determining the concentrations of gas are different between the manufacturer and the gas mixing system used by Jack. This ultimately leads to a 2-3% difference in the gas concentration. 
> Which is about the uncertainty of the premixed gas. Since the gas system was originally 75/25, but displays the premixed system as about 77/23 or 78/22, we originally saw the GEM efficiency lower even though the mixing system was operating properly. I 
> think this answers your question.
> 
> Zeke
> 
> On Mon, Nov 8, 2021 at 12:12 PM Bogdan Wojtsekhowski <bogdanw at jlab.org <mailto:bogdanw at jlab.org>> wrote:
> 
>     Looks nice. Could you summarize why the original use of the mixing system was
>     not successful?
>     ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>     *From:* Sbs_gems <sbs_gems-bounces at jlab.org <mailto:sbs_gems-bounces at jlab.org>> on behalf of Ezekiel Wertz <ewertz at email.wm.edu <mailto:ewertz at email.wm.edu>>
>     *Sent:* Monday, November 8, 2021 12:07 PM
>     *To:* sbs_gems at jlab.org <mailto:sbs_gems at jlab.org> <sbs_gems at jlab.org <mailto:sbs_gems at jlab.org>>
>     *Cc:* Jack Segal <segal at jlab.org <mailto:segal at jlab.org>>
>     *Subject:* [Sbs_gems] [EXTERNAL] Mixing System for GEMs
>     Hi GEM folks,
> 
>     I have attached a plot and will describe a discussion I had with Jack this morning pertaining to the GEM gas mix system. So if you have questions he may be able to address them better. I would like to preface that the mixing system is not connected to
>     the BigBite GEMs when these tests and manipulations were done.
> 
>     Recently (last couple of days) Jack was able to use the Binary Gas Analyzer, which I think analyzes the gas sample using a molecular technique. Jack would be able to provide a more accurate description of the technique. Whereas, the original mixing
>     system used a mass slow control based off of a voltage setpoint. What he noticed is that both systems agree pretty well.
> 
>     The plot pretty much summarizes the behavior of the GEM gas system, but I will provide a description as well. The original gas mixing system used for the GEMs was properly calibrated to 75/25 or 74/26 Ar/CO2. We see this early on in the plot (more to
>     the left). After this Jack then connected premix Ar/CO2 75/25, premix Ar/CO2 80/20, and 100% Argon. What we see is that with the Binary Gas Analyzer the premixed Ar/CO2 75/25 actually displays as closer to 77/23 or 78/22 for Ar/CO2. Then to the right
>     of the plot we see that Jack has calibrated the mixing system to be fairly identical to the premix gas value. In other words the mixing system is now calibrated to the same concentrations of Ar/CO2 as the premixed gas that the GEM folks are more
>     familiar with.
> 
>     So the gas system needs to flush for a couple of hours and it has been since about 10am this morning. From last week there is cosmic data using the premix system and GEM efficiencies are available. This afternoon we could switch to the mix system. Let
>     it flush the now calibrated mixed Ar/CO2 through the BigBite GEMs for at least 10 hours and then some time this week before returning to beam operations we could collect cosmic data and compare it to the efficiencies with the premixed gas.
> 
>     If there are any comments or questions please respond to this email.
> 
>     Best,
>     Zeke
> 

-- 
John(Jack) Segal
Halls A and C Spectrometer Support Manager
Thomas Jefferson National Accelerator Facility
12000 Jefferson Avenue
Suite 12
Newport News, VA 23606-4486
Phone: (757) 269-7242
Cell: (757) 320-9977