WEBVTT

1
00:00:04.930 --> 00:00:28.899
WH11NE- Sunrise: So, Hello, everybody. I'm at second year of Phd. At the School of Normal Superior, Pisa. I'm also associated with the National Institute of nuclear physics. And today I'm going to present the status and the unexpected applications of the so-called Pt modules for the outer tracker upgrade of Cms.

2
00:00:31.598 --> 00:00:33.092
WH11NE- Sunrise: It's working okay.

3
00:00:33.780 --> 00:00:37.310
WH11NE- Sunrise: So my plan for today is to

4
00:00:37.510 --> 00:00:42.680
WH11NE- Sunrise: try to answer this question that I'm sure everybody here

5
00:00:42.850 --> 00:01:10.799
WH11NE- Sunrise: ask all the time and keep up at night. So why Cms need a new tracker? What do we want to do with this tracker? What are those Pt modules that everybody keep telling about? Are we really building them. What were we gonna do with them? And as a last thing, what is this Muni experiment that is, gonna use this pt module

6
00:01:11.750 --> 00:01:21.020
WH11NE- Sunrise: so just a quick survey to to fine tune the the presentation who is from Cms here?

7
00:01:22.040 --> 00:01:31.449
WH11NE- Sunrise: Okay, basically everybody. So the 1st slide will be not so- so instructive because, we

8
00:01:31.530 --> 00:01:44.170
WH11NE- Sunrise: present the Cms experiment at Cern. So it's your common high energy physics detector designed to study proton-proton collision at

9
00:01:44.170 --> 00:02:05.380
WH11NE- Sunrise: 13 TV of center of mass energy at is the design is the common particle detective design. So it, we have a super conducting solenoids that generate a 3.8 Tesla field, and inside of it we have a track, a silicon tracking station

10
00:02:05.530 --> 00:02:29.369
WH11NE- Sunrise: that is used to measure the charge, and the momentum of charged particles as well as secondary vertex is coming from long decaying particles. Then we have the electromagnetic calorimeter and the hydronic calorimeter to measure the energy of electrons and photons and the hydrons, and then, outside of the

11
00:02:29.960 --> 00:02:42.319
WH11NE- Sunrise: of the superconducted solenoid, we have neon chambers to identify with the high purity the new particles

12
00:02:42.650 --> 00:03:05.768
WH11NE- Sunrise: since the collisions between problems happens at 49 Hertz. We cannot save all the data that we can collect. So we have a 2 stage system of triggers to select them of interest. The 1st one called the level one is based on hardware, so on and so forth. And he, he have a very low latency

13
00:03:06.550 --> 00:03:27.519
WH11NE- Sunrise: algorithm that are used to select event based only on the simplified information of the parameters and the new ones, while the second system. The high level trigger is based on software is, has a higher latency and is, use the information from all the detector.

14
00:03:28.650 --> 00:03:34.480
WH11NE- Sunrise: So another thing that I pretty sure everyone in this room knows is that

15
00:03:34.740 --> 00:03:57.740
WH11NE- Sunrise: the Lhc machine will have an upgrade in the next few years called high luminosity. Lhc. That will increase the instantaneous luminosity of the machine, and this will bring in the plan of the high luminosity an increase of the acquired statistic by one order of magnitude.

16
00:03:57.760 --> 00:04:20.669
WH11NE- Sunrise: And why? This is interesting? Because with the increased statistic we will have new opportunities for searching rare channels or making a precision measurement. Here I gave only 2 examples. The 1st one is this, so that the coupling of the heats to massive particles, and we see a projection

17
00:04:20.670 --> 00:04:29.679
WH11NE- Sunrise: do the 3 act inverse autobahn of the final acquired statistic of Highly. Let's see. And here

18
00:04:30.110 --> 00:04:32.670
WH11NE- Sunrise: we can see the improvement

19
00:04:32.820 --> 00:04:47.769
WH11NE- Sunrise: in the measurement of the red decay of the B sub s. Into that new ones with the with the precision, and circulated to the same of the statistic of a high luminosity. Let's see, and we can see that the the

20
00:04:47.950 --> 00:05:04.479
WH11NE- Sunrise: not only we, will be able to measure more precisely the piece of S into the new ones, but we expect also to have the sensitivity to measure for the 1st time at 5 c. Significance, the B. And to the new ones. But here I cheated

21
00:05:04.480 --> 00:05:18.600
WH11NE- Sunrise: because it's not only a thing of of having a more performance, collider and increase the statistic. But I, as you can see by the the shape of the of the of the peak

22
00:05:18.600 --> 00:05:30.529
WH11NE- Sunrise: here is narrow, so we expect also to have an improvement in the actual detector otherwise, and these will improve as well our set of measurements. But

23
00:05:31.060 --> 00:05:52.020
WH11NE- Sunrise: is this necessary? Actually, yes, because the I'm I will focus on this only on the tracker. But this this applies also for for other sub detectors. But regarding the tracker, our current tracker is designed for the specs of phase one so Lhc nothc.

24
00:05:52.020 --> 00:06:15.319
WH11NE- Sunrise: for example, the actual sensor can withstand the radiation damage of the high luminosity environment. And it also is not designed the current detector to deal with the high pile up. That is the number of primary collisions that happens at every bunch crossing here. We can see

25
00:06:15.320 --> 00:06:25.999
WH11NE- Sunrise: around from 2016 if I'm correct. Yes, that shows a pile up of 100. So each red point is a primary collision.

26
00:06:26.000 --> 00:06:39.279
WH11NE- Sunrise: and each yellow line is a track coming from one of these primary collisions. So it's not trivial to reconstruct and assign the track to the correct.

27
00:06:39.360 --> 00:06:58.040
WH11NE- Sunrise: to record the vertex, and this is with pile up of about 100. But we expect luminosity to have even worse conditions. So a pile up between 140 and 200 with respect to the average right now, which is 60,

28
00:06:59.800 --> 00:07:22.050
WH11NE- Sunrise: so what are, what is our plan to survive to this harsh environment? We need a new tracker. And since we are building this new tracker, our requirement. Our ideal detector should have this feature 1st of all, of course, radiation tolerance. That is enough to withstand the

29
00:07:22.070 --> 00:07:45.549
WH11NE- Sunrise: their integrated. Then we want the high granularity for having low Channel occupancy and having efficient tracking. We would like to have less material budget in the tracking volume to, of course, make peaks narrower. So having more resolution in the momentum and the invariant mass.

30
00:07:45.760 --> 00:08:08.530
WH11NE- Sunrise: and then a very cool feature that we would like to have is to have a to give a contribution to the 1st level of the of the trigger. So this so-called level one tracking to improve the selection, purity of the events of interest, and with this higher selection purity keep the level one, accept rate manageable

31
00:08:09.180 --> 00:08:31.619
WH11NE- Sunrise: without sacrificing too much the physics as an extra. We have also decided that our next level, one trigger, have more time for making his decision. So all these features are very important to the design of the new tracker.

32
00:08:33.320 --> 00:08:36.779
WH11NE- Sunrise: So let's talk about this pt module, and let's

33
00:08:36.919 --> 00:08:51.773
WH11NE- Sunrise: begin considering why we need a design which is not usual. So not the the Pt module little spoiler are not made by only one layer of silicon. So not only one silicon

34
00:08:52.320 --> 00:09:19.179
WH11NE- Sunrise: sensor. Why, here I show a very simple back on the envelope calculation. If we consider a reasonable pixel pixel sensor pixel model, with around 30,000 channels and an average channel of occupancy at the per 1 million level, which is the expected of occupancy in the pixel layers at high luminosity. Let's see.

35
00:09:19.200 --> 00:09:37.209
WH11NE- Sunrise: Well, yeah, we need at least 19 bits of information to for each for each hit. And we want to send this information via optical links to the backend boards in order to reconstruct tracks, and so on.

36
00:09:37.570 --> 00:09:58.519
WH11NE- Sunrise: is every every packet of data that has to be sent out via via optical links as a header, let's say about 30 bits is reasonable in which we want to to send the information of the branch crossing the status bits to monitor the sensor. So this simple

37
00:09:59.110 --> 00:10:14.060
WH11NE- Sunrise: The simple calculation brings our unexpected rate of 24 gigabits per second, which is not feasible for our fastest transceiver, the local gpt, which runs up 10 gigabits per second.

38
00:10:14.440 --> 00:10:34.460
WH11NE- Sunrise: And okay, someone could say, yeah, but you can, let's say, send clusters making very intelligent, very clever encoding of the information. But still this is not enough to give the a reasonable data rate.

39
00:10:34.680 --> 00:10:46.050
WH11NE- Sunrise: And also we don't want to reduce to, let's say, 10 GB per second. We want to reduce the data rate by one order of magnitude, just to be sure that also the

40
00:10:46.250 --> 00:10:53.069
WH11NE- Sunrise: event that we are efficient also sending event that are above the the average occupancy.

41
00:10:54.160 --> 00:11:04.030
WH11NE- Sunrise: So to solve the issue. The idea that Cms people have is to use the fact that the distribution of charged particles

42
00:11:04.410 --> 00:11:20.020
WH11NE- Sunrise: is, let's say, roughly exponential. So it's by by having even low. Pt. Cut here, let's say, at the 2 Gev. Which is the nominal cut, we we plan the

43
00:11:20.240 --> 00:11:38.020
WH11NE- Sunrise: in high luminosity. Hc. For the tracker. So even imposing a low Pt. Cut. So without sacrificing too much the physics, we can still reduce the rate by one order of magnitude, and so, having the data transmission manageable.

44
00:11:38.200 --> 00:11:43.520
WH11NE- Sunrise: how we we can do it well, using a pair of sensors

45
00:11:43.530 --> 00:12:06.430
WH11NE- Sunrise: close the spaces nearby each other, and we have this strategy. We have a bottom sensor support. Also seed layer, in which a charge particle pass and leaves energy, and we have a we have a hit here. So, starting from this seed, we construct a programmable search window

46
00:12:06.849 --> 00:12:29.949
WH11NE- Sunrise: in which we search for the other heat of the of the same truck. So if we find inside this correction, though it means that the the chart, the the truck, is is a high pity, so we can. We can send it outside of the we can send outside to the back end

47
00:12:30.160 --> 00:12:31.150
WH11NE- Sunrise: boards.

48
00:12:31.260 --> 00:12:53.839
WH11NE- Sunrise: If the track is a flow, Pt. The other hit in the other layer, the correlation layer will fall outside this search window, and then we reject the truck because it's it's a flow. Pt, and we are not interested in in this kind of tracks, and this is a way to make the rate manageable.

49
00:12:55.520 --> 00:13:22.459
WH11NE- Sunrise: So here we have the pity modules that implements this strategy of momentum selection on the front end we have the 2 s. Module, with 2 strips layers for the layer, the the seed layer, and the correction layer, both strips there, and the Ps module in which we have a Macro pixel sensor and a strips sensor.

50
00:13:22.690 --> 00:13:26.480
WH11NE- Sunrise: So let's focus of how these Pt modules are made.

51
00:13:26.960 --> 00:13:55.510
WH11NE- Sunrise: Let's start with the 2 s. Model. We have this 2 layer with a with a strip pitch of 90 micron that are not. This is a thing that I would like to underline is these 2 layers cannot be stereo layers, because otherwise we can lose the ability to to have this momentum reconstruction. So they are parallel to each other.

52
00:13:56.060 --> 00:14:13.939
WH11NE- Sunrise: Strips are made like these, and we have on each side a so called flexible hybrid. So it's a hybrid. That's that's the technology that enables us to make the correlation between the 2 sensors. So this flexible material

53
00:14:13.940 --> 00:14:43.879
WH11NE- Sunrise: that is used to read out the the 2 layers. In this case we have a a common front end chip, called the Cpc. Which reads out, both the layers, make the stop selection, and then sends its data to a concentrator. See chip one for each hybrid, and then these these 2 concentrators selects the the steps in case there are too many, and we cannot send every

54
00:14:44.090 --> 00:14:58.990
WH11NE- Sunrise: every study, the selects, the one with the highest 15, and then each hybrid, and each send the information to the transceiver and to the and then to the- the

55
00:15:00.040 --> 00:15:05.060
WH11NE- Sunrise: with the module. Things are slightly more complicated

56
00:15:05.120 --> 00:15:27.219
WH11NE- Sunrise: because we have the micro pixel sensor, which is bump bonded to this Mpa front end while the strict sensor is readout by wire bonding by this Ssa. Front end. So the Ssa. Sent the stream information to the Macro Pixel. Sorry to the Mpa.

57
00:15:27.220 --> 00:15:49.049
WH11NE- Sunrise: And the Mpa. Make the stop selection, and then, as in the in the 2 S. Case, they, the stops, are sent to the concentrator, chip one for each hybrid on the 2 sides of the of the module. And then there's a there's a 2 concentrator chip sent to the transceiver the the data.

58
00:15:50.660 --> 00:16:02.460
WH11NE- Sunrise: So here we have the a quarter of our outer tracker, so we have a region of a barrel made of 2 s. Module.

59
00:16:03.350 --> 00:16:05.799
WH11NE- Sunrise: An end. Cap d. Is committed of both

60
00:16:06.110 --> 00:16:28.000
WH11NE- Sunrise: 2 s. And Ps. Module, and we have the barrel made of Ps. Module, with 2 subsection, a flat barrel, and a tilted barrel. The tilted barrel is is there? For various reasons? One of them is that, of course, if you need to make a correlation between 2 layers, you don't want

61
00:16:28.070 --> 00:16:56.890
WH11NE- Sunrise: edging efficiency. So imagine having a flat barrel all over here most of the of the tracks at the edge of the of the detector will be in one sensor and in the other there will be not present, because, of course, they they are, they have an angle. And so this design helps in adding the efficiency of the of the selection of the stops. Better

62
00:17:00.250 --> 00:17:05.500
WH11NE- Sunrise: so 50 models have a have a

63
00:17:06.150 --> 00:17:33.219
WH11NE- Sunrise: a complex system of chips, and have a complex system of data stream, because we have 2 data streams. The 1st is the one I told right now. So the stabs data which are simplified data, the sense a reduced information of the correlated event in both sensors. So we have the heat in the heat layer plus a simplified version of the

64
00:17:33.220 --> 00:17:53.940
WH11NE- Sunrise: then, so that they offset of the second in the in the second layer, and this is sent at 40 megahertz rate to build the track primitives. So the front end. Send the stuff to the concentrator and then to the transceiver, and then on the back end. We have this data acquisition board that collects the the stamps

65
00:17:53.940 --> 00:18:05.099
WH11NE- Sunrise: from from all the models that are connected to this to this board, and sends to a fact finder processor. So the the board that is responsible for building the tracks.

66
00:18:05.100 --> 00:18:26.260
WH11NE- Sunrise: These tracks are sent as a primitive to the level, one trigger as well with the others of the data center formation, and after 12.5 microsecond the level, one trigger, decide if the if the event is worth acquiring or not, and then send the so called level one. Accept back

67
00:18:26.260 --> 00:18:53.103
WH11NE- Sunrise: to the front end board that distributes the the signal to all the modules in the front end. Chip the complete data from both of the the sensor. We're stored in a pipeline because we want to to- to send out the complete information only for level. One accepted the data. So the the data are collected from this pipeline and sent to the system, and then to the

68
00:18:54.610 --> 00:18:56.876
WH11NE- Sunrise: to the hey, to the.

69
00:19:00.310 --> 00:19:06.279
WH11NE- Sunrise: So just to mention it for completeness.

70
00:19:07.091 --> 00:19:24.409
WH11NE- Sunrise: I want to quickly show how the tracking works at level one we have the so called, which is based on the idea of having stamps to a couple of subs from

71
00:19:24.410 --> 00:19:45.689
WH11NE- Sunrise: a tablet from 2 different layers, which which are used to reconstruct a track, or to derive the parameters of a track. Assuming that this is a prompt track, not displaced track, so we construct the the parameters of the helix of the track, we project

72
00:19:45.860 --> 00:20:06.670
WH11NE- Sunrise: the the track to the other layer that we're not taking into account, and if we find stuff in the other layer we have the. So we have this matching. The track candidate is accepted for defeated via account

73
00:20:06.690 --> 00:20:22.439
WH11NE- Sunrise: a common filter, and then is also to the track is applied quality counter via Bdt. For sure that we have precise time track parameters and enhance purity of this sample of data.

74
00:20:23.000 --> 00:20:29.980
WH11NE- Sunrise: And this is done on Fpgas with a latency of 4 microseconds.

75
00:20:31.010 --> 00:20:49.419
WH11NE- Sunrise: Okay, the whole system. We design with the idea of having this efficiency high efficiency, helping the the level one trigger, so has it effect. The simulation says that it has actually effect.

76
00:20:49.510 --> 00:21:13.090
WH11NE- Sunrise: Here I show only an example with a single neuron trigger. We see that for a given threshold, the rate with the track information without so using only the new Chamber information, the rate is reduced a lot. So this is very, very useful to manage the rate of level one

77
00:21:13.150 --> 00:21:22.580
WH11NE- Sunrise: trigger, and we see that the turn on curve is is way sharper. So we have a much higher resolution of the track momentum

78
00:21:22.790 --> 00:21:25.640
WH11NE- Sunrise: upper level one. And this helps a lot.

79
00:21:27.340 --> 00:21:34.000
WH11NE- Sunrise: So this it was a let's say the theory. So the the idea behind the

80
00:21:34.640 --> 00:21:39.940
WH11NE- Sunrise: the Pd modules and the tracker upgrade. But Pt modules are

81
00:21:40.270 --> 00:21:50.980
WH11NE- Sunrise: starting to be built and tested. So here I would like to to show the the activities that we do in Pisa with this module.

82
00:21:51.120 --> 00:22:01.689
WH11NE- Sunrise: Pisa as well as Fermilab, is a so called burning and integration site. So the-the modules are built in assembly site.

83
00:22:02.050 --> 00:22:06.609
WH11NE- Sunrise: which, in our case our assembly site is body, and the new job.

84
00:22:07.150 --> 00:22:13.530
WH11NE- Sunrise: They are received a visa, and they are supposed burning text tested more on that later.

85
00:22:13.530 --> 00:22:40.029
WH11NE- Sunrise: And then, after they are properly tested, that everything in the model works. They are integrated in the tilt and the tilted barrel which I showed before. So here we have the the clean room pizza with the various rings from various layers of this bar, and here we see some dummy sensor put on place to test if the mounting actually works.

86
00:22:40.130 --> 00:22:57.059
WH11NE- Sunrise: So what is this burning box? And why is it interesting? Well, burning box is physically a box that was developed and each integration site? Yes, the trigger efficiency. Yeah.

87
00:22:59.560 --> 00:23:12.379
WH11NE- Sunrise: yeah. Sorry. Could you just explain the difference between the red and black and also the circle? Well, I guess the circle in box is the different data. Yeah, these are, these are different details. So the idea is that

88
00:23:12.770 --> 00:23:37.499
WH11NE- Sunrise: that, they are around 14, and they're on the simulation for the selection of a single neurons, using only the information of new chambers. And here's the rate of accept from the from here, when instead is integrated also the level one tracking. So we have the

89
00:23:37.950 --> 00:23:54.100
WH11NE- Sunrise: global new in a sense. So we have the-, the tracker and the chambers. We see that the rate is lower, and there's a even for low pity. So this is an enhancement in the selection. And here is to show that

90
00:23:54.300 --> 00:24:15.549
WH11NE- Sunrise: this is why this is possible, because the resolution on the Pt. Momentum is higher. So here we see the how many single neurons are accepted with a threshold of 20 Gb. Considering the actual through the Pt of the neuron.

91
00:24:15.580 --> 00:24:29.010
WH11NE- Sunrise: So we see that here we select also new ones that are not. They are below 20 Gv. Because of a resolution of the immune chamber is worse than considering both the new chambers and the tracker.

92
00:24:29.910 --> 00:24:31.518
WH11NE- Sunrise: Okay, cool. Thank you.

93
00:24:34.380 --> 00:24:39.030
WH11NE- Sunrise: So coming back to to the burning box is

94
00:24:39.420 --> 00:25:08.769
WH11NE- Sunrise: a box, and why is it in? Is interesting? Because this is a this box provides thermal and light insulation for testing the modules, and provides all the services that are needed to run and operate the modules. So low voltage for powering high voltage, for the sense of depletion, cooling, dry air, and also the optical result.

95
00:25:08.990 --> 00:25:31.310
WH11NE- Sunrise: And this is a key step in the qualification of the modules, because when we receive the module, we do this order of 100 thermocycle between room temperature and minus 35°C, which which will be the the operating temperature of the of the tracker in phase 2,

96
00:25:31.310 --> 00:25:42.419
WH11NE- Sunrise: and these we-, we do these to test the model mechanics, and also the electronics of the of the model to test if it works. And if it works, we integrate in the.

97
00:25:42.830 --> 00:25:44.289
WH11NE- Sunrise: in, the, in the rings.

98
00:25:44.680 --> 00:25:49.740
WH11NE- Sunrise: But looking at this picture, you see that this is actually

99
00:25:50.030 --> 00:25:58.150
WH11NE- Sunrise: a great setup for acquiring cosmic data because we have a stack, actually, 2 stacks of modules

100
00:25:58.320 --> 00:26:19.619
WH11NE- Sunrise: with the all the services provided. So that can fully function. We have a data position board, which is not show this here almost yet if we extended the photo. So we have this board that can read out the modules so we can use it as a cosmic, great great telescope.

101
00:26:19.690 --> 00:26:43.570
WH11NE- Sunrise: And in Pisa we we implemented that this function of cosmicane. This is still a work in progress. There's other work to be done. But this is a very interesting application, because, in a sense the module itself triggering, because when the 2 layers are fired from

102
00:26:43.810 --> 00:26:52.039
WH11NE- Sunrise: from a from a charged particle. These triggers the send or stop, so we can receive

103
00:26:52.040 --> 00:27:13.839
WH11NE- Sunrise: the steps, and we can select a coincidence of the steps to to build events from cosmic Ray. Here we see the. We see the correlation between the position of the of the heat in the here we tested only with the 2 links, because the

104
00:27:13.910 --> 00:27:17.079
WH11NE- Sunrise: there are not that many modules around

105
00:27:17.470 --> 00:27:22.480
WH11NE- Sunrise: at the moment. So with only 2 links, we see that there is a correlation, so that then

106
00:27:23.920 --> 00:27:38.189
WH11NE- Sunrise: we see that the distribution of of course we Greece is almost as we expect, all, I say, almost because we expect more tilted

107
00:27:38.990 --> 00:27:55.550
WH11NE- Sunrise: cosmic rays. But since we have the stop selection, the all these tilted rays are rejected. So we have a sharper distribution, and here we see the the coincidence in time between the 2 layers is for the

108
00:27:55.550 --> 00:28:18.640
WH11NE- Sunrise: 2 hybrids of the of the same link, of the of the same module with respect to the other. So we see that these we, we prove that in principle this can work. So there's still work to be done to make it full functioning. But this opens new possibilities for testing more aspects of the model, the efficiency of the module.

109
00:28:18.690 --> 00:28:32.749
WH11NE- Sunrise: We can test a new new alignment algorithm in this simple setup. We can test the align, the timing alignment of the model. So it's very interesting to making more tests with this with this model.

110
00:28:33.120 --> 00:28:44.089
WH11NE- Sunrise: But actually, pity models reveal that themselves as very flexible models, because they are used also for this so called Muni experiment

111
00:28:44.250 --> 00:28:45.120
WH11NE- Sunrise: cool

112
00:28:47.650 --> 00:29:14.049
WH11NE- Sunrise: here, because in a nutshell I will try to explain a thing that probably most of you already know, so that in 2023 the firmware experiment Mu, one g. Minus 2, published the most precise measurement of this positive mu, one anomalous magnetic moment. And this result is in tension with the standard model.

113
00:29:14.050 --> 00:29:25.870
WH11NE- Sunrise: Yeah, we found new physics. Actually, no, we don't know. Actually, because the standard model prediction, the theoretical standard model prediction, is not so trivial to be done.

114
00:29:26.340 --> 00:29:39.019
WH11NE- Sunrise: because in the standard modification of G minus 2. We have various contribution. The the 1st 2 terms from the Qad. And the weak force are very precisely known

115
00:29:39.020 --> 00:30:00.990
WH11NE- Sunrise: known, so we know that very well. Instead, the contribution from Qcd. Is is known with a large incentive. We have this, the main contribution with the so-called hydronic vacuum polarization, and the other. I don't remember the acronym, and we see that the

116
00:30:01.120 --> 00:30:19.059
WH11NE- Sunrise: the pretty, the theoretical prediction made with the lattice calculation, data driven approaches are not in. Are very. They have a lot of uncertainty, and they don't agree with each other

117
00:30:19.300 --> 00:30:24.999
WH11NE- Sunrise: very well. So the idea of Muni is to measure this contribution.

118
00:30:25.450 --> 00:30:51.809
WH11NE- Sunrise: The hydronic vacuum polarization directly using elastic scattering of muons into electrons. So how you can determine that is adronic vacuum polarization. We use the differential cross section of this elastic scattering, because here we have the differential Cross section. We can determine that the term by integrating this

119
00:30:51.890 --> 00:31:12.820
WH11NE- Sunrise: this form one and the Differential Cross section is related to measuring the angle of scattering. So we have the the scattering angle from the electron and the new one. And here, by measuring very precisely this, this cube. We can obtain the this term.

120
00:31:12.940 --> 00:31:23.639
WH11NE- Sunrise: but how we can do it we we need very good tracking to do it. So, and many people decided to use the 2 S modules to do it. So

121
00:31:24.510 --> 00:31:41.930
WH11NE- Sunrise: so the design of the experiment is within this, so called the tracking station. We have many of them to increase the perception of the interaction. Each station has its own target, and it has 6 to s. Modules.

122
00:31:42.110 --> 00:32:00.059
WH11NE- Sunrise: which are used, of course, to tracking the the scattering event. Then the experiment will have a nickel to to measure the energy of the electron and also a neon chamber of some sort. Then to to identify them.

123
00:32:01.010 --> 00:32:16.570
WH11NE- Sunrise: So how the the 2 S models are are placed. We have 2 arrangements. We have the Xy layers which are tilted with respect to the, to the, to the beam, to increase the probability of charge, sharing

124
00:32:17.057 --> 00:32:39.469
WH11NE- Sunrise: for data, and thus increasing the resolution of the of the measurement of the position, and then we have the so called UV layer to to resolve the the reconstruction ambiguities that can happen with more than one track, using only only Xy layers.

125
00:32:40.510 --> 00:33:07.200
WH11NE- Sunrise: So we, the result is of only the stops data at 40 megas. And we are the the new experiment will have online building on Fpga, using the the same board that user. And they are also working on online tracking. And so this is a very interesting concept for using the Pt. Modules.

126
00:33:08.220 --> 00:33:16.750
WH11NE- Sunrise: So in the last 2 years there there have been a certain, an extensive testing campaign. Actually, this is me

127
00:33:16.770 --> 00:33:25.999
WH11NE- Sunrise: from 2023, and we. We run the test beam at the M. 2 beam

128
00:33:26.000 --> 00:33:48.910
WH11NE- Sunrise: at Cern, which is a beam of new ones at 160 Gb. In which this is the great difference with respect to the Cms environment. The new ones are not in time with some, so they're not separated by 25 microsecond, which is the same

129
00:33:49.020 --> 00:33:58.679
WH11NE- Sunrise: that but they are they. They arrive asynchronously in a 5 second spill. So this is a a different environment.

130
00:33:58.830 --> 00:34:17.520
WH11NE- Sunrise: We use the 2 or 3 graphite target between the 2 tracking stations. So the 1st station is used to measure the direction of the incoming neurons, and then the second station is used to reconstruct the vertices and the elastic scattering of the

131
00:34:17.710 --> 00:34:22.139
WH11NE- Sunrise: of the of the new one over the elections of the graphite target.

132
00:34:22.570 --> 00:34:37.750
WH11NE- Sunrise: due to lack of of modules. The 3rd session was empty, but the idea in the 1st complete run that will be 2025 is to populate also. This 3rd session of 2 S. Module to make new identification.

133
00:34:38.350 --> 00:34:42.839
WH11NE- Sunrise: During this testing campaign we were able to collect

134
00:34:43.060 --> 00:34:54.429
WH11NE- Sunrise: around the 300 TB of steps data, and also Carlo data. And so we collected the order of 10 to the 8 elastic scattering events.

135
00:34:55.460 --> 00:35:18.689
WH11NE- Sunrise: And here are the preliminary results that are part of the of the proposal of phase one. So we see that the tracking performance is in line with the expectation. We see here the residual of the of the the resolution of the of the 2 s. Module, made with the residual between the the

136
00:35:18.690 --> 00:35:26.120
WH11NE- Sunrise: position of the heat by the truck feet and the actual position of the sensor, and we see that

137
00:35:26.190 --> 00:35:32.390
WH11NE- Sunrise: there is some change in line with the with the expectation of a 90 micron feature module.

138
00:35:33.230 --> 00:35:36.329
WH11NE- Sunrise: Here we see the this

139
00:35:36.390 --> 00:35:58.560
WH11NE- Sunrise: standard deviation of the difference between the direction, the the direction, the angle direction of the track in the 1st station, and in the second station, and this is a measure on how how accurately we can align the station, how much impact the

140
00:35:58.560 --> 00:36:14.459
WH11NE- Sunrise: the multiple scattering. And so here we have the data without target and with targets, and you see that without target. Of course, as expected, we have less difference in the angle.

141
00:36:14.760 --> 00:36:43.720
WH11NE- Sunrise: and then we see that we can actually reconstruct the vertex position, using the information of the second station. Here we can see the primary peak which is the graphite target, but we can also see the elastic scattering made in the 1st station of the of the 1st module of the second station, and in the last 2 layers of the 1st station, that

142
00:36:44.060 --> 00:36:48.260
WH11NE- Sunrise: once in a while they they behave like a target.

143
00:36:48.780 --> 00:37:02.889
WH11NE- Sunrise: This is for the, for the tracker performance, and here for the physics, performance for physics results. We see a 1st sign of the shape that we want

144
00:37:02.890 --> 00:37:17.690
WH11NE- Sunrise: to measure. And actually, this is at least to me very impressive. This is the shape and the direction of the beam spot reconstructed using online tracking. This is impressive because

145
00:37:17.690 --> 00:37:29.770
WH11NE- Sunrise: 2 S. Pt. Modules are designed to do tracking at 40 Megahertz online. And this is the 1st time that they, they, we see that they are actually able to do it.

146
00:37:29.770 --> 00:37:54.209
WH11NE- Sunrise: Okay, the environment is simpler because we don't have a magnetic field. We don't have the bilap of Cms, but still we are able to reconstruct tracks at 40, Megahertz, and the plan is also to do vertexing at 40 Megahertz. So last year a phase, one proposal was submitted for uni to request 4 weeks of data, taking in 2025,

147
00:37:54.440 --> 00:38:07.790
WH11NE- Sunrise: with the idea of estimating the 1st contribution of the G minus 2, so the electromagnetic one, just as a sanity check that the whole system works.

148
00:38:09.370 --> 00:38:21.840
WH11NE- Sunrise: So, in conclusion, I hope to have answered this very, very super interesting question. So, and I I would.

149
00:38:22.000 --> 00:38:35.923
WH11NE- Sunrise: I hope that I showed you that these modules that were designed for a very, very specific environment, very, very specific task actually showed that they are indeed very flexible, and they can use in a very

150
00:38:37.830 --> 00:38:58.942
WH11NE- Sunrise: Why, the K, why the use cases. And and this is a pity concept of this as a some days ago. Detector in a module concept works very well. So that's all for me. And thanks for the attention

151
00:39:02.460 --> 00:39:03.420
WH11NE- Sunrise: or something.

152
00:39:04.880 --> 00:39:29.460
WH11NE- Sunrise: Okay, okay, do we have questions from the audience or on Zoom? Also, I didn't catch like, what is it about the experiment that requires

153
00:39:29.660 --> 00:39:36.081
WH11NE- Sunrise: like this- this new tracker? Okay, the idea is that traditional

154
00:39:37.271 --> 00:39:45.850
WH11NE- Sunrise: the idea is that you have to collect a very high statistics. And this

155
00:39:46.620 --> 00:40:07.960
WH11NE- Sunrise: in in the, in the ideal scenario, Muni people will have to design. And this is a thing that is foreseen for a phase. 2 muni. They have their own silicon tracking detectors fine tuned for their needs, but at the time that experiment was proposed the available detectors

156
00:40:08.120 --> 00:40:16.980
WH11NE- Sunrise: out there from the all the available detectors. The only one that suited enough the needs of this this experiment was the the Pt. Modules.

157
00:40:17.070 --> 00:40:28.980
WH11NE- Sunrise: so there were no other detectors that were not able that they were able to withstand the rate that they were expecting actually doesn't need the the Pt concept.

158
00:40:28.980 --> 00:40:48.530
WH11NE- Sunrise: They they even thinking about a a sort of setup with one S sensor. So with only one strip to reduce the material budget to the multiple scattering. But at the time. This is the only modules that is able to do it, and actually is the only one that is able

159
00:40:48.530 --> 00:41:07.417
WH11NE- Sunrise: to send the data. At 40 megas by themselves without sending a trigger signal. So they are able to this. This, actually, this 1st selection is very helpful for doing online tracking online vertexing. And this is

160
00:41:07.890 --> 00:41:12.560
WH11NE- Sunrise: this is very useful for this kind of online selection.

161
00:41:19.860 --> 00:41:31.180
WH11NE- Sunrise: So are you gonna use like Cms to S modules, or are you gonna build new ones? I there's a no there are like

162
00:41:31.270 --> 00:41:59.669
WH11NE- Sunrise: Cms. Granted promised. I don't know the the politically correct, the term to say this, so promised around the 300 modules or so to populate all the the stations, and at least for 2025 promised, like 18 order of 20 modules from pre-production and prototype modules to be used in the

163
00:41:59.950 --> 00:42:03.340
WH11NE- Sunrise: in the setup, in the in the 4 weeks that are taking.

164
00:42:04.000 --> 00:42:09.020
WH11NE- Sunrise: So there! There are deals. I I don't know that the politics behind, but they are like.

165
00:42:10.740 --> 00:42:20.269
WH11NE- Sunrise: but they are the same kind. It's like the same design, same. It's it's a 2 s module. You can use it. And in Cms is they are basically the same.

166
00:42:20.910 --> 00:42:26.700
WH11NE- Sunrise: They're gonna borrow them for some time, and then you will have

167
00:42:27.430 --> 00:42:32.480
WH11NE- Sunrise: your set. No, I think they are. They will have their own set. Yes.

168
00:42:34.730 --> 00:42:50.399
WH11NE- Sunrise: I had another question about just the this scattering. So the physics of it. So you are using that target? A graphite target? Yeah. But you're interested in the scattering with electrons. So what about the rest of the.

169
00:42:50.810 --> 00:43:08.710
WH11NE- Sunrise: So that's a that's a background. Like, yeah, you you use only the the target for the electrons of the shell from the from. They are thinking also from target if I'm correct right now, the I think for the for the

170
00:43:08.720 --> 00:43:24.700
WH11NE- Sunrise: that has been the only available or the only useful was the graphites, but they are thinking about very new. And so the the one of the main background will be the scattering of neurons into nuclei that will

171
00:43:25.700 --> 00:43:44.799
WH11NE- Sunrise: produce other products. So this is, this will be a source of background. But they are only interested in in the, in the electrons from the material. So so basically, they try to get rid of the showers. And just yeah, the point of this is just to get.

172
00:43:46.680 --> 00:44:12.829
WH11NE- Sunrise: and they have many, many stations, because you put an M. Yeah, the idea is 40. The design is 40, and except for the 1st one, the other one has a target. This is to increase the the cross section and the statistics of elastic scattering, and it is supposed to be assigned with this, yeah, attend to beam, which is the the beam of compass.

173
00:44:13.254 --> 00:44:18.599
WH11NE- Sunrise: Think you will be in front of compass? Or maybe they I don't remember but that there will be the the.

174
00:44:19.070 --> 00:44:21.560
WH11NE- Sunrise: the ideal, the ideal meeting.

175
00:44:21.690 --> 00:44:22.510
WH11NE- Sunrise: Yes.

176
00:44:24.270 --> 00:44:37.610
WH11NE- Sunrise: another question that maybe, is not so. Just a curiosity, a thinking of what we will do. So you mentioned that in Pisa you had the possibility to mount them in the tilted barrel.

177
00:44:38.040 --> 00:44:54.659
WH11NE- Sunrise: Configuration right with the module, I guess, to populate. But but you have the setup. Yeah, right? So so one thing is, I imagine it's gonna be a challenge for the

178
00:44:55.130 --> 00:45:06.460
WH11NE- Sunrise: does. Does the fact that it's a tilted borrow challenge the way you set up the programmable

179
00:45:06.710 --> 00:45:10.089
WH11NE- Sunrise: bt for for the triggering

180
00:45:10.140 --> 00:45:15.839
WH11NE- Sunrise: right? So the angle at which or is it? Is it like you say, Okay, is there like

181
00:45:15.840 --> 00:45:42.860
WH11NE- Sunrise: angle the the modules there at a certain angle. They all use this, the ones that are closer, you know. Higher, Ito or no, you you can. You can program. It is how much is that it affects? I don't know specifically for for rings, but you have. The the window is programmable in 2 way. The width of the of the window, so you can accept the depending on the region of the tracker you are.

182
00:45:42.990 --> 00:45:44.879
WH11NE- Sunrise: Where is the clock

183
00:45:45.760 --> 00:46:04.000
WH11NE- Sunrise: depending? If we, if you're here, there, and depending on the spacing, because there will be more spacing like between 1.6 and 4 will be the situation. So you can. You can fine tune the the width to optimize, for the the nominal cut is a 2 gb.

184
00:46:04.170 --> 00:46:20.739
WH11NE- Sunrise: so you you can- you can tune the- the window to- to have the card right in all the regions of the of the tracker, and you can also tune the the offset of the window. So here in the

185
00:46:21.120 --> 00:46:38.030
WH11NE- Sunrise: the example here, the the search window is like symmetrically respected to the, to the seed layer. But you can also offset a bit. And actually, this is the thing that is doing so in the- in the tilted layer.

186
00:46:39.950 --> 00:46:41.359
WH11NE- Sunrise: Oh, Price.

187
00:46:42.430 --> 00:47:03.459
WH11NE- Sunrise: yeah, this tilted layers, the the configuration of the model is done in such a way that you you don't lose efficiency for for the scattering that opposite in the direction with respect to the tilting, so you offset to the correlation window a bit to be straight. With respect to the to the beam

188
00:47:06.820 --> 00:47:24.109
WH11NE- Sunrise: seems a very complicated yeah, yeah, this is very complicated. That's gonna be. It's it's a it's an open thing. I think the the line they're working on your knee, because also in your knee, you have another issue that if you want to measure angles.

189
00:47:24.530 --> 00:47:48.110
WH11NE- Sunrise: you need to know precisely this spacing, but also the longitudinal distance between the modules and this is without a a magnetic field is a weak mode of the of the tracking station, so the the idea is to use a particular event with the secondary vertices in different start in different targets. By

190
00:47:48.580 --> 00:48:15.839
WH11NE- Sunrise: and with this you should be able to reduce the the issue of the the weak mode. And also you have this metal logic measurement. At least this is the thing for for muni. They use also the the inver frame, which is a very particular alloy that has a very low thermal expansion. Coefficient. They they flux the whole

191
00:48:16.180 --> 00:48:28.919
WH11NE- Sunrise: chamber with the air at a given temperature, just to be sure that the thermal fluctuation is possible. But yes, this is complicated in various.

192
00:48:29.130 --> 00:48:45.429
WH11NE- Sunrise: The the price you have to pay to have this Pt selection is that okay? It's very cool. You can also select the model. But the model gets very complicated from an electric from electronic point of view, from a configuration point of view, and also from an alignment point of view.

193
00:48:45.960 --> 00:48:58.289
WH11NE- Sunrise: but the I mean the stop is for the selecting the events. But then, if the event is selected, we have all the heat information

194
00:48:58.410 --> 00:49:01.500
WH11NE- Sunrise: already, like stuff that is on a cylinder.

195
00:49:01.830 --> 00:49:24.160
WH11NE- Sunrise: I don't know. It seems complicated. And now now you also have all of these different angles at different distance, and it seems like a big, big problem. Well, big, I mean, it's not solvable. But I guess that way more complicated than than the current track the other question that it was about this about the trigger is, how many stubs

196
00:49:24.380 --> 00:49:30.964
WH11NE- Sunrise: you get with one module like so when you get like 200 or something piled up.

197
00:49:31.390 --> 00:49:55.069
WH11NE- Sunrise: okay, I was sure that this kind of question will appear. So this is, just to show how the the data packet is sent out from the Cic. So the concentrator chip, and you have this, the status of the various contents and the number of subs and depending of the of the

198
00:49:55.070 --> 00:50:07.639
WH11NE- Sunrise: flavor of the electronics you can have at most 19 for the Ps. Stops for the after 19. After

199
00:50:09.050 --> 00:50:24.630
WH11NE- Sunrise: 5 Giga gigabit per second. You have twice the number for 10 gigabit per second the front end, and for the 2 s. You have at most 16 stops, but also there are

200
00:50:24.840 --> 00:50:34.540
WH11NE- Sunrise: very characteristic modes where you can. You can also send the data without the information, and you can acquire more

201
00:50:35.637 --> 00:50:58.350
WH11NE- Sunrise: more stuff. I don't think at the moment it is independent in the fitting procedure. So technically, you can go without a stop.

202
00:50:58.620 --> 00:51:12.740
WH11NE- Sunrise: but and something that you can program. So you can decide on the spot the only thing with respect to the number that you said that this is for 8, so every single is 1 8th of that

203
00:51:13.310 --> 00:51:27.219
WH11NE- Sunrise: or well, it's not that we can have a bunch of way more than other with less. Yeah, this is for integrated at 8, because you can have events with more steps. So you use a a packet of 8 bunch of

204
00:51:28.630 --> 00:51:30.769
WH11NE- Sunrise: to integrate between

205
00:51:36.043 --> 00:51:49.699
WH11NE- Sunrise: momentum measurement possible. Yes, there's no physics price to pay for not having them stereographed like you.

206
00:51:50.500 --> 00:51:58.230
WH11NE- Sunrise: Of course you have the the the price to pay that you don't get the data coordinate, but

207
00:51:58.970 --> 00:52:09.702
WH11NE- Sunrise: I mean you. You still have 3 layers of Pixel Street, the Ps. Modules. So you have that information, and also the

208
00:52:10.330 --> 00:52:28.479
WH11NE- Sunrise: at the At level one. You you're trying to reconstruct only prompts at. Then on the you get information also from the inner tracker that has with the pixels, with even a higher resolution. So you don't pay that much.

209
00:52:29.570 --> 00:52:36.680
WH11NE- Sunrise: Can you go to the plot we showed the sure if it's for the too much.

210
00:52:45.341 --> 00:52:47.869
WH11NE- Sunrise: this one. Yeah, this is the.

211
00:52:51.440 --> 00:52:57.360
WH11NE- Sunrise: This selection is applied to reconstructed nuance.

212
00:52:58.000 --> 00:53:00.780
WH11NE- Sunrise: It's like this doesn't show, like the

213
00:53:01.590 --> 00:53:07.770
WH11NE- Sunrise: as the efficiency to have it be reconstructed in submission at all right. This is just showing the momentum smear.

214
00:53:08.286 --> 00:53:11.060
WH11NE- Sunrise: Yeah, this is-, this is showing that the

215
00:53:11.827 --> 00:53:19.010
WH11NE- Sunrise: like. If I take all the if I take all the events below 20 gb, do they perfectly fill in that little space underneath the

216
00:53:19.180 --> 00:53:32.387
WH11NE- Sunrise: underneath, like one? No, no, no, these are these are only the ones selected with this, this threshold of 20 there, not all the the new ones below 20. Gb, okay.

217
00:53:33.520 --> 00:53:37.590
WH11NE- Sunrise: these are only the ones selected with when you apply a 20 Gb. Threshold.

218
00:53:40.470 --> 00:53:53.070
WH11NE- Sunrise: Okay, so maybe stupid. But let's say, my is there any way that, like the the

219
00:53:53.680 --> 00:53:57.424
WH11NE- Sunrise: like, the physics triggers suffer with the

220
00:53:58.630 --> 00:54:15.110
WH11NE- Sunrise: stone paradigm like, is there like, if you like, the be like meanders out, and then sends out 2 new ones that are kind of forward and like right underneath one of the layers, and it'll it'll fly outwards. Sort of parallel ish to the

221
00:54:15.300 --> 00:54:16.660
WH11NE- Sunrise: the sensors.

222
00:54:16.850 --> 00:54:28.980
WH11NE- Sunrise: So you see, like for rejection, because they are too close, and they and they're flying out. Honestly, I don't know the effects of stops for for

223
00:54:30.773 --> 00:54:42.469
WH11NE- Sunrise: physics. The only thing I know they are trying to get it is to have the the chocolate algorithm, not only for prompt tracks, but also for displaced tracker.

224
00:54:42.690 --> 00:54:53.369
WH11NE- Sunrise: For up to 10 cm of displacement to improve the because yeah. Here, yes, the you assuming from tax

225
00:54:54.800 --> 00:55:01.839
WH11NE- Sunrise: you lower the efficiency for this place track, of course, but this is the bottleneck here is that

226
00:55:02.200 --> 00:55:09.989
WH11NE- Sunrise: you have to run it on Fpga. So right now that the baseline algorithm, they they were able to make it work.

227
00:55:10.270 --> 00:55:23.589
WH11NE- Sunrise: So now they have to integrate, and then they will try to- to extend to triplets, not stop, not doublet of the subs to use 3, and you retrospect the

228
00:55:24.210 --> 00:55:28.359
WH11NE- Sunrise: now going to use the same.

229
00:55:28.672 --> 00:55:29.650
WH11NE- Sunrise: I don't know if they want.

230
00:55:29.960 --> 00:55:30.770
WH11NE- Sunrise: Yeah.

231
00:55:30.920 --> 00:55:39.699
WH11NE- Sunrise: but these still only forget the ones that are offline to trigger on, and then they can just get it.

232
00:55:44.910 --> 00:55:51.970
WH11NE- Sunrise: So if there are. No, there are no other pressing questions. I think we can thank Marco and.