sPHENIX MAPS tracker discussion

[Anthony Frawley <afrawley AT fsu.edu>]

Hello Xin,

The layer phi misalignment is important, as demonstrated by the fact that it greatly increases the width of the measured collision vertex distribution. If you use the cluster positions provided by the offline tracking software there is a correction in the default alignment parameter files for the stave rotation misalignment. It is just Yasser's correction to the phi0 for each layer.

The MVTX has to be moved in (x,y) from its nominal position in sPHENIX by about (+6.2 mm, -2 mm) to put the beam spot at about (0,0) in the sPHENIX coordinate system. That is really all that can be said from collision data.

I don't think we have any evidence that the tracking performance will be affected by the displacement of the MVTX relative to the beam spot. The tracking software makes no assumption about how well centered the detectors are. The only effect would come from increased cluster density at some phi angles, but I doubt that it would be enough to affect the pattern recognition efficiency.

Best regards

Tony

---

From: sphenix-maps-l-request AT lists.bnl.gov <sphenix-maps-l-request AT lists.bnl.gov> on behalf of Xin Dong <xdong AT lbl.gov>
Sent: Friday, July 5, 2024 12:13 PM
To: Cameron Thomas Dean <cameron.dean AT cern.ch>
Cc: josborn1 AT bnl.gov <josborn1 AT bnl.gov>; Xudong Yu <yuxd AT stu.pku.edu.cn>; SPHENIX MVTX <sphenix-maps-l AT lists.bnl.gov>
Subject: Re: [[Sphenix-maps-l] ] [Sphenix-maps-l] [EXTERNAL] Re: MVTX meeting next Monday 7/1 12:30PM/ET

 

Hi Cameron,

Thanks for the clarification. Then I think it is clear from the phi oscillation that the beam position is not centered in MVTX (as you stated in your previous email, but I was a bit confused by the 4mm bump you described before). And I would request we need to shift the beam orbit to the center of MVTX instead of sPHENIX. The phi dependent cluster density distribution would introduce an azimuthal dependence tracking performance, this will introduce quite some complication in the downstream analysis e.g. tracking efficiency, dca resolution etc. The best is to flatten the azimuthal angle dependence of MVTX cluster distribution in operation. I didn't check the beam pipe constraints at injection Cameron mentioned. Hopefully this can be worked out.

The calculation doesn't depend on the mis-alignment calibration much. It only depends on the stave azimuthal position. For this study,  the MVTX cluster phi position needs to be calculated assuming MVTX sits at (0,0). The global mis-alignment of MVTX w.r.t the sPHENIX center shouldn't be relevant here, but certainly, we want to be clear if the software is handling it consistently. For stave position, we are talking about the dependence at the scale of full 2pi range, so tens of mrad rotation also shouldn't have a large impact. This was seen in the offline QA comparisons.

Thanks and Best Regards

/xin 

On Fri, Jul 5, 2024 at 8:35 AM Cameron Thomas Dean <cameron.dean AT cern.ch> wrote:

Hi Xin,

For the cluster positions, the plots shown should be with the ideal position which assumes the MVTX is centered around RHIC (0,0). We can use an alignment file with the offset in it which has been used before but I don’t think was used here.

As a clarification, the 4mm bump I mentioned is a physical change in the beam orbit that C-AD had to implement to minimize any problems from the displaced beam pipe.

Cheers,
Cameron

 

From: sphenix-maps-l-request AT lists.bnl.gov <sphenix-maps-l-request AT lists.bnl.gov> on behalf of Xin Dong <xdong AT lbl.gov>
Date: Friday, July 5, 2024 at 11:21 AM
To: josborn1 AT bnl.gov <josborn1 AT bnl.gov>
Cc: Xudong Yu <yuxd AT stu.pku.edu.cn>, Cameron Thomas Dean <cameron.dean AT cern.ch>, SPHENIX MVTX <sphenix-maps-l AT lists.bnl.gov>
Subject: Re: [[Sphenix-maps-l] ] [Sphenix-maps-l] [EXTERNAL] Re: MVTX meeting next Monday 7/1 12:30PM/ET

Hi Joe,

 

Thanks for the information. This is as I expected. The rotation in MVTX layers we are talking about is very small (0.0X rad), compared to the extracted phi0 (which is about 3). The impact on the (X0,Y0) is negligible. Actually the first several runs in the figure were from HaoRen with filename labeled p001, and there is no visible difference in the X0,Y0 parameter.

 

To Cameron, thanks for the material regarding beam positions. Actually I do have a question which needs you or others to clarify. In the offline QA plot, the MVTX cluster positions, are they calculated assuming MVTX sits at (0,0) or it already includes the 4mm global shift you are referring to? I was told it was handled as the ideal case (which I take as natural before any mis-alignment calibration is finalized), but your message seems to indicate the later one.

 

Thanks and Regards

 

/xin

 

 

On Fri, Jul 5, 2024 at 4:39 AM Osborn, Joseph <josborn1 AT bnl.gov> wrote:

To Xin - any file you looked at with "p004" in the filename should include a fix for the layer rotation. I would also agree with Cameron that it is important that we separate the effects from the beam spot to the actual misalignment of the MVTX barrel from the nominal origin of sPHENIX.

 

Joe

 

___________________________

Joe Osborn, Ph.D

Physics Department

Brookhaven National Laboratory

josborn1 AT bnl.gov

---

From: sphenix-maps-l-request AT lists.bnl.gov <sphenix-maps-l-request AT lists.bnl.gov> on behalf of Cameron Thomas Dean <cameron.dean AT cern.ch>
Sent: Friday, July 5, 2024 7:22 AM
To: Xin Dong <xdong AT lbl.gov>; Xudong Yu <yuxd AT stu.pku.edu.cn>
Cc: SPHENIX MVTX <sphenix-maps-l AT lists.bnl.gov>
Subject: Re: [[Sphenix-maps-l] ] [Sphenix-maps-l] [EXTERNAL] Re: MVTX meeting next Monday 7/1 12:30PM/ET

 

Hi all,

Sorry, there may have been some confusion from our discussion on Monday. The beam is centered in sPHENIX. The MVTX is not centered in sPHENIX, and is in fact tilted. We had many discussions with C-AD about the position of the MVTX and how that effects the orbit. There’s already a 4mm bump to account for the incorrect installation of the MVTX.

 

https://indico.bnl.gov/event/23005/

https://docs.google.com/document/d/1WJAGBi-m3w8hljgJKcrj-Ix8n3zJOmIB72V9Php-Er8/edit?usp=sharing

Remember, the beam at injection is already near the limit of the beam pipe due to our installation.

Cameron

 

From: sphenix-maps-l-request AT lists.bnl.gov <sphenix-maps-l-request AT lists.bnl.gov> on behalf of Xin Dong <xdong AT lbl.gov>
Date: Thursday, July 4, 2024 at 1:18 PM
To: Xudong Yu <yuxd AT stu.pku.edu.cn>
Cc: SPHENIX MVTX <sphenix-maps-l AT lists.bnl.gov>
Subject: Re: [[Sphenix-maps-l] ] [Sphenix-maps-l] [EXTERNAL] Re: MVTX meeting next Monday 7/1 12:30PM/ET

Hi Xudong,

 

Thank you very much for digging into this in detail. Your MC test is quite informative and offers quite new insights.

 

Hi All,

 

I attach the latest run-dependent beam line center distributions. They look quite stable. Ming, would you bring this up to the operation team and we should make the request to CAD to shift the beam orbit.

 

Thanks and Best Regards

 

/xin

 

 

On Wed, Jul 3, 2024 at 11:08 PM Xu-Dong Yu <yuxd AT stu.pku.edu.cn> wrote:

Dear all,

 

According to the private discussion with Xin, i claim that there is no problem in Xin’s calculation. There are some conclusion

 

1. my derivation is correct, but it is only for 1D case. Stave has finite length in z direction, so we should also consider z-dimensional solid angle coverage. The Jacobian in z or theta direction is the same factor of 1+d/R cos(phi-phi0) which will make this term squared. Due to small quantity of d/R, we can expand it into 1+2d/R cos(phi-phi0) which is Xin’s formula. If stave is infinite length, it will reduce to 1D case without factor of 2. This has been proved by MC simulation.

 

2. The hits/clusters distribution on z dimension is important. We have tried different emitting particle eta/theta distribution which makes hits/clusters z distribution different from that in data. The result will deviate from the true value according to the MC simulation. In data, z distribution is flat, which ensure then density in different z is the same, and will keep the formula with factor of 2.

 

3. A example MC simulation is shown below. Three plots are attached.

 

  1) eta distribution is assumed to be [exp(-eta) + eta(eta)] / 2 and eta region is [-3,3]

 

  2) hit z distribution in layer 2 which is uniform same with that in data

 

  3) Fit to the cluster phi distribution w.r.t. mvtx center. The fit formula is Xin’s with factor of 2. The input value of d is 4mm, and the output value of phi0 is 3rad. The output is consistent with input.

 

4. In software, mvtx center is placed to sPHENIX center (0,0,0), and we only use single mvtx detector information, we can not extract MVTX global translation information. Therefore, my argument about alignment is wrong.

 

Sorry to bother those who aren't interested in the details.

 

Best regards,

 

Xudong

 

 

On Jul 3, 2024, at 01:08, Xin Dong <xdong AT lbl.gov> wrote:

 

Hi Xudong,

 

The distribution is plotted as a function phi, but for all the clusters integrated over the full z region. While all staves have the same z length, so it will be a constant scaling factor compared to the hit area density distributions on the cylinder surface.

 

dN/dz will not be a constant as it also depends on the distance to the beam position while this distance scales with the radial distance (rphi direction and z direction are both orthogonal to the radial direction). Thus, the area density dependence will need to be squared then. For clarification, one may run a MC to test it out.

 

Feel free to ping me offline if you prefer more detailed discussions.

 

Thanks and Best Regards

 

/xin 

 

 

On Tue, Jul 2, 2024 at 8:06 PM Xudong Yu <yuxd AT stu.pku.edu.cn> wrote:

Dear Xin,

 

Thanks for your reply. I don't understand why you said "area density", since you fit to 1D phi distribution. Isn't it only a 1D pdf w.r.t. phi? Even if you consider z distribution, it should be dN/dz=1?

For second question, I mean (-8mm, 2mm) is the position of beam position w.r.t. MVTX center,  and MVTX center is not the same as sPHENIX center. As fas as i know, due to installation, MVTX center is about 5mm west to the sPHENIX center.

 

MVTX center x + (-8mm) = beam x position ≈ vertex x position. And we know that, according to Hao-Ren's vertex distribution, vertex peak at x=-5mm. Therefore, i said it is consistent with vertex distribution. If i am wrong, please correct me.

 

Best regards,

Xudong

Original:

• From:Xin Dong < xdong AT lbl.gov>
• Date:2024-07-03 10:28:01
• To:yuxd < yuxd AT stu.pku.edu.cn>
• Cc:SPHENIX MVTX < sphenix-maps-l AT lists.bnl.gov>
• Subject:Re: [Sphenix-maps-l] [EXTERNAL] Re: MVTX meeting next Monday 7/1 12:30PM/ET

Hi Xu-Dong,

 

Thanks for looking into this. The hit density we are looking at here is the area density. At the leading order, there will be another same factor (1+ d/R * cos(phi-phi0)) for dN/dz, therefore, the hit density we are looking at here would be (1 + d/R *cos(phi-phi0)) ^2.  Keeping the leading term, it will become 1 + 2*d/R*cos(phi-phi0). I think you may also run a MC to test it.

 

I have one question regarding your statement "... (-8mm, 2mm), ... which is consistent with vertex distribution peak position." What do you mean here? And also MVTX center is not necessary to be the sPHENIX center.

 

Thanks and Best Regards

 

/xin

 

 

 

On Tue, Jul 2, 2024 at 6:30 PM Xu-Dong Yu <yuxd AT stu.pku.edu.cn> wrote:

Dear Xin,

 

Thanks for your very nice and very interesting study. I have a comment that maybe your formula has an extra coefficient of 2. After correcting it, beam position w.r.t. MVTX center should be about (X0,Y0) ~ (-8mm,2mm) which is consistent with vertex distribution peak position which is w.r.t. sPHENIX origin. You can see the attached manuscript. If i am wrong, please refute me. Thanks

 

Best regards,

 

Xudong

 

 

 

On Jul 2, 2024, at 16:35, Xin Dong via sPHENIX-MAPS-l <sphenix-maps-l AT lists.bnl.gov> wrote:

 

Hi Joe and HaoRen,

 

Thanks a lot. The runs after 47002 contain the MVTX cluster phi distributions. The summary of the run dependence can be found below.

<MVTX_BL_run.png>

I see there are more runs appearing in the directory. I can keep updating the figure with more data coming in.

 

Thanks and Best Regards

 

/xin

 

On Tue, Jul 2, 2024 at 12:16 PM Osborn, Joseph <josborn1 AT bnl.gov> wrote:

Hao-Ren is correct. Those files are being aggregated, they are much larger so they take longer to aggregate. But they show up in the same directory under *CLUSTER* instead of *SEED*.

 

 

 

___________________________

Joe Osborn, Ph.D

Physics Department

Brookhaven National Laboratory

josborn1 AT bnl.gov

 

 

---

From: Hao-Ren Jheng
Sent: Tuesday, July 2, 2024 3:02 PM
To: Xin Dong
Cc: Osborn, Joseph; SPHENIX MVTX
Subject: Re: [Sphenix-maps-l] [EXTERNAL] Re: MVTX meeting next Monday 7/1 12:30PM/ET

 

Hi Xin, all,

 

I think the MVTX cluster QA module is already merged into the coresoftware: https://github.com/sPHENIX-Collaboration/coresoftware/blob/master/offline/QA/Tracking/MvtxClusterQA.cc

Newer runs should have the cluster histograms from the production.

 

Best,

Hao-Ren

 

On Jul 2, 2024, at 14:19, Xin Dong via sPHENIX-MAPS-l <sphenix-maps-l AT lists.bnl.gov> wrote:

 

Hi Joe,

 

Thanks. I took a look at several root files. It seems the MVTX cluster QA histograms (cluster vs. phi particularly, not associated with tracking, just all clusters) are not saved in these files. Probably we should encourage Hao-Ren to integrate his code updates into the production so we can monitor them along with the run goes.

 

Thanks and Best Regards

 

/xin

 

On Tue, Jul 2, 2024 at 11:05 AM Osborn, Joseph <josborn1 AT bnl.gov> wrote:

You can find many runs produced as a part of the production here:

/sphenix/data/data02/sphnxpro/QAhtml/aggregated/*SEED*

 

Take a look and let me know if you find issues,

 

Joe

 

___________________________

Joe Osborn, Ph.D

Physics Department

Brookhaven National Laboratory

josborn1 AT bnl.gov

---

From: sPHENIX-MAPS-l <sphenix-maps-l-bounces AT lists.bnl.gov> on behalf of Xin Dong via sPHENIX-MAPS-l <sphenix-maps-l AT lists.bnl.gov>
Sent: Tuesday, July 2, 2024 1:58 PM
To: Liu, Ming Xiong <mliu AT lanl.gov>
Cc: SPHENIX MVTX <sphenix-maps-l AT lists.bnl.gov>
Subject: Re: [Sphenix-maps-l] [EXTERNAL] Re: MVTX meeting next Monday 7/1 12:30PM/ET

 

Hi Ming and All,

 

I looped through all the runs that Hao-Ren generated (actually turns out only 7 runs have the valid entries). Please see the updated slides in the same indico location

https://indico.bnl.gov/event/24069/contributions/93529/attachments/55585/95138/20240701_MVTX_BeamLine.pdf

 

The run dependence can be found here. It is reasonably stable up to 46677. 

<MVTX_BL_run.png>

 

Hi Haoren, if you have new production data, feel free to share. 

 

I also put the macros and figures in the following Git directory.

https://github.com/starsdong/MVTX/tree/main/2024/BeamLine

 

Comments are very welcome. Thanks and Regards

 

/xin

 

On Tue, Jul 2, 2024 at 8:44 AM Liu, Ming Xiong <mliu AT lanl.gov> wrote:

Hi Xin and Hao-Ren,

 

Nice! 

Yes, I agree we should have the beam centered wrt to MVTX for best physics, and we  

 

Hao-Ren, did you use the ideal geometry or the updated one from Yasser’s corrections?  

 

Cheers,

Ming  

 

 

-- 

Dr. Ming Xiong Liu

P-3, MS H846

Physics Division

 

Office: 505.667.7125

Mobile: 505.412.7396

Los Alamos National Laboratory

 

 

From: Xin Dong <xdong AT lbl.gov>
Date: Monday, July 1, 2024 at 11:59 PM
To: Hao-Ren Jheng <hrjheng AT mit.edu>
Cc: Ming Liu <mliu AT lanl.gov>, SPHENIX MVTX <sphenix-maps-l AT lists.bnl.gov>
Subject: [EXTERNAL] Re: [Sphenix-maps-l] MVTX meeting next Monday 7/1 12:30PM/ET

 

Hi All,

 

Thanks Hao-Ren for generating the QA histograms. Based on his root file, I did a quick analysis and please see the findings below

 

https://indico.bnl.gov/event/24069/contributions/93529/attachments/55585/95095/20240701_MVTX_BeamLine.pdf

The slides are also posted to today's MVTX meeting indico page

 

The proposed actions can be found in the last page:

 

1) W.r.t. the condition for the data that Hao-Ren generated, I would propose we request CAD to move the beam orbit at sPHENIX IR horizontally towards x+ direction by 4mm, and if possible, also vertically downwards by 1mm.

 

2) But, the beam position may change, particularly considering recent changes in terms of beam conditions, we should re-analyze the latest beam data and check again.

 

3) We should continue monitoring the beam position throughout the Run. The MVTX cluster phi distributions are a good measure to monitor this.

 

Comments and suggestions are more than welcome. Thanks and best Regards

 

/xin

 

 

On Mon, Jul 1, 2024 at 8:46 PM Xin Dong <xdong AT lbl.gov> wrote:

Hi Hao-Ren,

 

Thank you. I have one question regarding the MVTX cluster phi distribution. You labeled "w.r.t origin". So what origin was used here for the cluster phi angle calculation? If you can clarify, that would be much appreciated.

 

Thanks and Best Regards

 

/xin

 

On Mon, Jul 1, 2024 at 2:41 PM Hao-Ren Jheng <hrjheng AT mit.edu> wrote:

Hi Xin,

 

I just uploaded my slide to the indico page:https://indico.bnl.gov/event/24069/contributions/93529/attachments/55585/95090/MVTXComm-20240701.pdf.

You can find one of the histogram files under my working directory: /sphenix/user/hjheng/sPHENIXRepo/macros/TrackingProduction/HIST_PHYSICS_run2pp_new_2024p001-00046667-0000.root

(You may find the histograms in this file are not the same as the plots I showed in the meeting. I figure a run without any stave missing is more convenient for you to make use of it.)

 

Please let me know if you have any further suggestion to the QA module.

 

Best,

Hao-Ren

 

On Jul 1, 2024, at 16:22, Xin Dong via sPHENIX-MAPS-l <sphenix-maps-l AT lists.bnl.gov> wrote:

 

Hi Hao-Ren,

 

Could you post your nice QA slides you showed today at the MVTX meeting to indico? Particularly, if you can also share with me the histogram root files for the MVTX layer cluster density distributions vs. phi and the event Vy vs. Vx, I can try to do a quick calculation on the beam position shift w.r.t to the MVTX center and we may feed back to the operation team and make the request.

 

Thanks and Regards

 

/xin

 

 

On Fri, Jun 28, 2024 at 8:39 AM Liu, Ming Xiong via sPHENIX-MAPS-l <sphenix-maps-l AT lists.bnl.gov> wrote:

Dear all,

We will have MVTX meeting next Monday at the usual time, 12:30PM/ET.

·                MVTX operation status 

·                Offline analysis- raw data QA & monitoring, noise/threshold, alignment etc.

·                AOB

 

Draft agenda, https://indico.bnl.gov/event/24069/

 

Join ZoomGov Meeting
https://bnl.zoomgov.com/j/1614542233?pwd=VExKVlVPaWZBM0JYaGhHb1EzcDluUT09

Meeting ID: 161 454 2233
Passcode: 430884
One tap mobile
+16692545252,,1614542233#,,,,*430884# US (San Jose)
+16692161590,,1614542233#,,,,*430884# US (San Jose)

 

 

Cheers,

Ming 

 

-- 

Dr. Ming Xiong Liu

P-3 Group, MS H846

Nuclear and Particle Physics & Applications

Physics Division

 

Office: 505.667.7125

Mobile: 505.412.7396

Los Alamos National Laboratory

_______________________________________________
sPHENIX-MAPS-l mailing list
sPHENIX-MAPS-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/sphenix-maps-l

 

 

--

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================

_______________________________________________
sPHENIX-MAPS-l mailing list
sPHENIX-MAPS-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/sphenix-maps-l

 

 

 

--

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================

 

 

--

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================

 

 

-- 

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================

 

 

-- 

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================

_______________________________________________
sPHENIX-MAPS-l mailing list
sPHENIX-MAPS-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/sphenix-maps-l

 

 

-- 

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================

_______________________________________________
sPHENIX-MAPS-l mailing list
sPHENIX-MAPS-l AT lists.bnl.gov
https://lists.bnl.gov/mailman/listinfo/sphenix-maps-l

 

 

--

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================

Error! Filename not specified.

 

--

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================

 

 

--

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================

 

--

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================

--

==================================================
Xin Dong
Nuclear Science Division, Lawrence Berkeley National Laboratory
MS70R0319, One Cyclotron Road
Berkeley, CA 94720, USA
Tel: +1-510-486-4121
Email: XDong AT lbl.gov
==================================================