1
00:00:01,439 --> 00:00:04,169
Right, we see your talk, Mark, please go
ahead.

2
00:00:04,799 --> 00:00:05,699
Hey, still see it?

3
00:00:07,140 --> 00:00:08,280
Yes, looks good.

4
00:00:08,880 --> 00:00:15,060
Oh, my pleasure to present the status of
the LCP upgrades. So as the overview of

5
00:00:15,060 --> 00:00:19,050
the talk, I'll start with a bit of
motivation about why the flight physics in

6
00:00:19,050 --> 00:00:24,120
the island of Seattle to be interested in,
I'll go through the various upgrade

7
00:00:24,120 --> 00:00:29,730
program that we have. So at least if we
run one, and run two we collected and

8
00:00:29,850 --> 00:00:33,510
integrate luminosity around 10 and plus
from two bands, and now we're going to

9
00:00:33,510 --> 00:00:39,300
upgrade first. Our first upgrade is to
increase the instantaneous luminosity by a

10
00:00:39,300 --> 00:00:45,300
factor of five. Now we have to cope with a
pileup that was one in in the first

11
00:00:45,300 --> 00:00:50,670
iteration of LGB. And then we got a pileup
of five and then this will run through

12
00:00:51,120 --> 00:00:57,330
bump three and one four of the, of the
Elysee and then during LS three, we'll

13
00:00:57,360 --> 00:01:02,490
make some small incremental improvements.
To, for example, the electromagnetic

14
00:01:02,490 --> 00:01:08,370
calorimeter we have to replace the animal
modules because of radiation damage. And

15
00:01:08,370 --> 00:01:12,360
then we now start to plan for a second
upgrade, but we would increase the

16
00:01:12,360 --> 00:01:17,340
instantaneous luminosity by a factor of 10
and try to run with a much, much higher

17
00:01:17,340 --> 00:01:23,130
pileup again. So we would have about 50
interactions, the crossing and the data

18
00:01:23,130 --> 00:01:27,600
said that we would have at the end of at
the end of the upgrade one would be 50

19
00:01:27,600 --> 00:01:33,450
inverse femto bonds and at the end of the
second upgrade, we would like to click the

20
00:01:33,930 --> 00:01:39,060
300 investment of bonds. Okay, so there
are a lot of talks from Alex up. So

21
00:01:39,060 --> 00:01:43,020
results physics results from like should
be one and then also some talks on the

22
00:01:43,020 --> 00:01:48,870
upgrade and you can find these throughout
the conference. So why LTV? I hope max in

23
00:01:48,870 --> 00:01:56,190
his talk this morning gave. Explain why
while it should be interesting, but it's a

24
00:01:56,220 --> 00:01:59,820
dedicated heavy flavor experiment,
experiment the Elysee and the core part of

25
00:01:59,820 --> 00:02:05,070
it program was to measure a CP violation
in the B sector and study a B and C have

26
00:02:05,070 --> 00:02:11,070
drawn the case. And you can see in the
plot at the bottom which has the direction

27
00:02:11,070 --> 00:02:14,880
that the two B's be part of the beepers
that are produced and then they'll say

28
00:02:14,880 --> 00:02:18,990
they're produced either in forward or
backward direction. So, because of the

29
00:02:18,990 --> 00:02:22,980
nature of the production of some of these
beam reasons, that a lot of things happen

30
00:02:22,980 --> 00:02:29,460
in the loops we can make indirect searches
for for new physics. And during runs one

31
00:02:29,460 --> 00:02:32,970
or two then we extended our physics
program which most we make measurements in

32
00:02:32,970 --> 00:02:38,850
electroweak physics, direct searches, also
heavy ions fixed pocket stuff, so lacp we

33
00:02:38,850 --> 00:02:43,920
can consider it as a forward as a general
purpose detector in the forward region.

34
00:02:43,950 --> 00:02:50,460
You can see here the acceptance in
compared to the CMS experiment. So we had

35
00:02:50,460 --> 00:02:55,260
a few hints for new physics and I think
these flavor anomalies were discussed this

36
00:02:55,260 --> 00:03:01,410
afternoon and at least to be summary talk,
there's no discovery but I chose Why, at

37
00:03:01,410 --> 00:03:06,000
least he is interested in why studying
playdough physics gives you a different

38
00:03:06,000 --> 00:03:11,460
handle on searches for new physics galaxy,
then we would like to increase our

39
00:03:11,460 --> 00:03:15,120
dataset. So we can start to look at
extremely rare the case or we can start to

40
00:03:15,120 --> 00:03:20,220
look with the case with branching ratios
less than 10 or minus nine for example.

41
00:03:22,170 --> 00:03:26,130
And then from the from the hardware point
of view, during the run more than two,

42
00:03:26,130 --> 00:03:32,010
then we had a level zero hardware trigger
which limited the output rate to one takes

43
00:03:32,010 --> 00:03:37,620
the input data at 40 megahertz and then
limits the output rate to 1.1 megahertz.

44
00:03:38,340 --> 00:03:44,520
So we could try to increase the
instantaneous luminosity. But when we do

45
00:03:44,520 --> 00:03:49,320
this, the essentially if you look at the
plots we increase the instantaneous

46
00:03:49,320 --> 00:03:54,630
luminosity for the hydronic modes when we
cut harder and harder on transverse energy

47
00:03:54,630 --> 00:03:58,830
and transverse momentum and we don't
really gain anything in statistics. So the

48
00:03:58,830 --> 00:04:04,230
trigger yield kind of flattens All fire as
for the atomic mode, okay, we could just

49
00:04:04,230 --> 00:04:11,190
increase the velocity as we want then
because of the higher instantaneous

50
00:04:11,190 --> 00:04:17,610
luminosity then we degrade the detector
performance. And we also have to take into

51
00:04:17,610 --> 00:04:21,540
account the radiation damage if the
sensors, especially in the silicon

52
00:04:21,540 --> 00:04:28,740
sensors. So this was the performance
during Romana Memorial two. So, the things

53
00:04:28,740 --> 00:04:32,640
that are crucial for the physics program
and ECB is we need to separate the primary

54
00:04:32,640 --> 00:04:37,980
and secondary vertices with a very high
precision we need to be able to measure

55
00:04:37,980 --> 00:04:42,240
the decay time with a very high
resolution. And then also we need

56
00:04:42,840 --> 00:04:51,450
excellent resolution so that we have good
mass resolution. And then we also have the

57
00:04:51,480 --> 00:04:57,240
particle ID detectors so you can see the
chunko detect some apart from a chunk of

58
00:04:57,240 --> 00:05:02,250
the injectors have a different momentum.
We can separate right between proton k on

59
00:05:02,310 --> 00:05:03,570
pion neons

60
00:05:05,220 --> 00:05:09,810
and then in Rome on them we had this
trigger that was running them to be able

61
00:05:09,810 --> 00:05:12,780
to separate between electronic and
hydronic final states and this will be a

62
00:05:12,780 --> 00:05:19,230
complete software trigger in in upgrade
one. So, this performance is the benchmark

63
00:05:19,230 --> 00:05:27,480
for all of our upgrades. So, one core part
of our program is as I said is

64
00:05:27,600 --> 00:05:33,690
understanding the CPM mechanism and the CP
violation and this comes in standard model

65
00:05:33,690 --> 00:05:41,160
in CPM matrix. So, you can see here
already in our letter of intent, this was

66
00:05:41,160 --> 00:05:48,660
the, the limits on the CPM parameters in
the Wolfram Stein parameterization is

67
00:05:48,660 --> 00:05:55,680
Roland Roland eater. And then if you
import the CPM matrix is unitary So, you

68
00:05:55,680 --> 00:06:00,930
can impose relations like this, which then
can be displayed in this set complex plane

69
00:06:00,990 --> 00:06:06,690
as these angles alpha, beta and gamma and
the so called unitarity triangle. So this

70
00:06:06,840 --> 00:06:10,890
with all of the inputs from the B
factories from A to B and from other

71
00:06:10,890 --> 00:06:16,350
experiments, you can see this is the
status of the seeking metric ck

72
00:06:16,350 --> 00:06:22,890
measurements at in 2019. And after the
phase one, you can see the improvement. So

73
00:06:22,890 --> 00:06:28,230
this would include measurements from Bell
two from Atlas and from CMS. And as one

74
00:06:28,230 --> 00:06:34,980
example on this angle on the angle gamma,
then we improve the improve the precision

75
00:06:34,980 --> 00:06:43,170
for by an order of magnitude from now to
the end of our upgrade program. Then,

76
00:06:43,410 --> 00:06:48,270
we're also interested in extremely rare
the case. So, this is a summary of all of

77
00:06:48,270 --> 00:06:52,860
the major but the core measurements that
we expect to make and we see how it

78
00:06:52,860 --> 00:06:58,620
compares a bit. We have complimentary
measurements to build to and to Atlas. And

79
00:06:58,620 --> 00:07:04,020
there's one example from the radio case
you can look at bs to me or Btw, and we'd

80
00:07:04,020 --> 00:07:07,980
be able to measure it with a high
precision for the first time. The ratio

81
00:07:07,980 --> 00:07:13,500
between these two things and this gives us
some test of minimal flavor violation. So

82
00:07:13,500 --> 00:07:17,730
currently we have a measurement, that's a
relative error of 90%. And after the

83
00:07:17,730 --> 00:07:25,770
second upgrade, this will be reduced
around 10%. Okay, so now on talk mostly

84
00:07:25,770 --> 00:07:30,450
about the detectors, this was a long story
that started already in 2008, when we had

85
00:07:30,450 --> 00:07:33,690
the first expression, expression of
interest for an upgrade, and I'm through

86
00:07:33,690 --> 00:07:38,220
all of the TDR program from the different
sub detectors. So the conditions the

87
00:07:38,220 --> 00:07:42,750
instantaneous luminosity is a factor of
five what we had in one one or two, and we

88
00:07:42,750 --> 00:07:46,680
aim to click the total integrate
luminosity 50 inverse into bonds with this

89
00:07:46,680 --> 00:07:50,550
much higher pileup. And the challenge
obviously, is to maintain the

90
00:07:50,550 --> 00:07:53,940
reconstruction performance apart like the
performance and everything in this much

91
00:07:53,940 --> 00:08:00,420
other environment and to remove completely
the level zero trigger and Without the

92
00:08:00,420 --> 00:08:07,260
complete detector at 14 megahertz This is
the sketch of the detector. So, we replace

93
00:08:07,260 --> 00:08:11,220
completely the tracking system or the new
vertex detector and then new tracking

94
00:08:11,220 --> 00:08:16,680
stations before and after the magnet, then
we have the two chunk of detectors where

95
00:08:16,890 --> 00:08:22,020
the entire rich one is essentially
replaced and rich Do we have new photon

96
00:08:22,020 --> 00:08:28,080
detectors and then for the color emitters
in the neuron system, the hardware is

97
00:08:28,080 --> 00:08:32,040
essentially the same, but then we just
replace all of the electronics and then

98
00:08:32,040 --> 00:08:35,520
all of the detectors have to replace the
readout electronics such that we can run

99
00:08:35,520 --> 00:08:42,180
with 40 megahertz readout. So, for the
data processing, this is essentially the

100
00:08:42,180 --> 00:08:46,710
scheme that was used in round one and
round two, but this will be used again for

101
00:08:46,710 --> 00:08:52,710
the for the upgrade. So we take the LCB
bunch crossing we have a collision rate 40

102
00:08:52,710 --> 00:08:56,820
megahertz we have the detector readout,
which is running at 40 megahertz. Then we

103
00:08:56,820 --> 00:09:03,450
have to process the essentially the 30
megahertz and non empty events. And this

104
00:09:03,450 --> 00:09:10,110
is done on hlt farm. So we do a
reconstruction, partial reconstruction and

105
00:09:10,110 --> 00:09:15,720
we make a selection based on the
transverse energy transfer of men and

106
00:09:15,720 --> 00:09:20,430
things like this. And then after we select
these events, these events get referred to

107
00:09:20,430 --> 00:09:26,220
a desk where a subset of the data is used
to run alignment and the calibration

108
00:09:26,220 --> 00:09:30,990
procedure. And then the alignment
calibration is fed back into the

109
00:09:30,990 --> 00:09:37,860
reconstruction. And then we start the
process the the data that's on the disk is

110
00:09:37,860 --> 00:09:44,070
we do the full reconstruction. So already
in our high level, second level trigger,

111
00:09:44,160 --> 00:09:49,680
we're using the offline data quality. So
we have the offline reconstruction Roman

112
00:09:49,680 --> 00:09:55,140
already in this hltv two one, so we don't
have to do so much data processing

113
00:09:55,140 --> 00:09:55,590
offline.

114
00:09:57,929 --> 00:10:01,079
So these are the same alignment
coefficients. They can put back in the

115
00:10:01,079 --> 00:10:06,029
offline processing. And from this, we then
reduce the data rate down such we're right

116
00:10:06,029 --> 00:10:12,869
now at Hubert per second. And we have two
data streams. One is the full data stream,

117
00:10:12,869 --> 00:10:17,339
which then gets processed offline. And
then we have the second model, where we

118
00:10:17,339 --> 00:10:22,619
already have directly the physics objects
that come from the hlt. And we throw away

119
00:10:22,619 --> 00:10:28,769
the rest of the event so we don't store
the roar event. So we can keep a larger

120
00:10:28,769 --> 00:10:34,229
fraction of the events. And then one last
one recent development is that the full

121
00:10:34,289 --> 00:10:43,739
hlt one reconstruction will run on GPS. So
this is a sketch of the gun line. So the

122
00:10:43,739 --> 00:10:48,719
detector from an electronics This is all
on the ground. This is then data is

123
00:10:48,719 --> 00:10:53,759
transmitted to the data center on the
surface, but we have the PCs that do the

124
00:10:53,999 --> 00:11:00,839
building. And we have the boards which do
the data processing. So take That takes

125
00:11:00,839 --> 00:11:05,789
the data from the front and backs it makes
the event makes the event fragments which

126
00:11:05,789 --> 00:11:09,689
are then built on this event builder PCs
with the event builder network that you

127
00:11:09,689 --> 00:11:13,979
can see here. And then the data is
transmitted to this filter farm, which are

128
00:11:13,979 --> 00:11:21,149
CPUs when we process the data. So the
readout boards are this PCI 40 board which

129
00:11:21,149 --> 00:11:26,759
is also used by others. And it's a large
FPGA and can process data with up to 48

130
00:11:26,759 --> 00:11:31,409
links. And this has three different
firmwares that we use one for the controls

131
00:11:31,409 --> 00:11:36,299
one for the time in the distribution, the
clock, and one for the data acquisition,

132
00:11:36,989 --> 00:11:43,559
and then the GPUs. So in event builder
PCs, we have two slots where we put the

133
00:11:43,589 --> 00:11:53,099
GPUs in this. So the full hlt one can run
on a brown 500 GPUs. Now the vertex

134
00:11:53,099 --> 00:11:59,489
detector, this is the detector that's
closest to the interaction region. So like

135
00:11:59,489 --> 00:12:04,679
the color detector that we have it's
retractable so we can open when the Elysee

136
00:12:04,679 --> 00:12:12,389
object and then we close when the galaxies
in the physics motoring collided mode and

137
00:12:12,809 --> 00:12:16,949
so, we move much closer now to the beam.
So, the first measurement point is already

138
00:12:16,949 --> 00:12:23,879
a 5.1 millimeters whereas before it was
eight millimeters the detectors themselves

139
00:12:24,089 --> 00:12:28,439
you can see the modules here they operate
in a secondary vacuum and you have these

140
00:12:28,439 --> 00:12:33,809
aluminum foils, which separate the
detector volume from the beam volume. So,

141
00:12:33,809 --> 00:12:41,939
you can see you see them here on a thin
aluminum foils they are machined to 250

142
00:12:41,939 --> 00:12:48,029
microns thick, and then we do a chemical
process to actually come down on 50

143
00:12:48,119 --> 00:12:55,559
microns. And here you can see easy the
modules so in total are 5200 pixel modules

144
00:12:55,559 --> 00:13:04,949
with 41 million pixels to cover their
smaller, small area Now, you can see a

145
00:13:04,949 --> 00:13:10,889
module. So, the module was then. So, the
consensus you have four. So, the consensus

146
00:13:12,149 --> 00:13:17,789
module two on the side, two on the other
side and these are bonded bonded to the

147
00:13:17,819 --> 00:13:24,449
reader, a six which is 256 by 256 pixel
array. So, develop x. So, you have 12 of

148
00:13:24,449 --> 00:13:31,259
these per module, the set on a silicon
substrate which has a micro channels

149
00:13:31,289 --> 00:13:38,279
etched into this and then yeah, then we
use a battery of co2 cooling through the

150
00:13:38,309 --> 00:13:42,989
micro channels. So, that we can maintain
it maintain temperature of around minus 20

151
00:13:43,049 --> 00:13:50,999
degrees Celsius. So, we also have to deal
with high bandwidth. So, the hottest AC

152
00:13:51,209 --> 00:13:56,669
can transmit data at the order of 20
gigabits per second and the sensors

153
00:13:56,669 --> 00:14:02,129
themselves and op sensors we have to use
Extremely radiation how technology because

154
00:14:02,129 --> 00:14:08,309
the expected Fluence in an office region
is around eight times 10 to the minus 13

155
00:14:08,519 --> 00:14:14,639
neutron equivalent. So the second part of
the tracking system is the upstream

156
00:14:14,639 --> 00:14:21,029
tracker. So this is four layers of silicon
which upstream of the magnet compared to

157
00:14:21,029 --> 00:14:25,349
the previous detector has a much much
finer granularity and you can see here as

158
00:14:25,349 --> 00:14:33,719
you get closer to the beam, we go from
wide pitch sensors down to narrow narrow

159
00:14:33,719 --> 00:14:39,689
pitch sensors and in the innermost region,
then we have half size sensors with a

160
00:14:40,559 --> 00:14:44,789
small pitch and we have this cut off
region around where the beam pipe goes to

161
00:14:44,789 --> 00:14:47,309
the we improve the acceptance, hater.

162
00:14:49,049 --> 00:14:53,579
The modules themselves they're mounted on
a state so this is a state a sketch of the

163
00:14:53,579 --> 00:14:58,709
state and it's double sided. You can see
one side here and we have the seven

164
00:14:58,709 --> 00:15:03,629
silicon sensors on this side. sermon on
the other side and then read out for the

165
00:15:03,629 --> 00:15:11,279
new basic, which is the salt basic, which
takes analog data to six as a six bit ADC

166
00:15:11,279 --> 00:15:16,529
we do the pedestal and common mode
subtraction, zero suppression things here

167
00:15:16,529 --> 00:15:24,149
the sensors bonded to the to the basic
basics then bonded to these flex cables on

168
00:15:24,149 --> 00:15:28,079
the flex cables and then connected via the
so called picture turned on the flex

169
00:15:28,079 --> 00:15:33,389
circuit to the reader electronics which is
mounted outside the acceptance on the

170
00:15:33,419 --> 00:15:34,169
detector frame

171
00:15:35,190 --> 00:15:36,480
Mark about five minutes

172
00:15:37,260 --> 00:15:43,530
okay. So then we have the sim laying fiber
tracker. So these are long synthetic

173
00:15:43,530 --> 00:15:48,120
fibers which are read out with a silicon
photomultiplier at one end to Silicon

174
00:15:48,360 --> 00:15:54,780
multiply sit in these cold boxes, which
are because the sap ends up to be cool to

175
00:15:54,780 --> 00:15:59,820
around minus 40 degrees to cope with the
neutron radiation that's expected in the

176
00:15:59,820 --> 00:16:05,340
end upgrade, we have a new, a sick, which
is m read out. And then we do some

177
00:16:05,340 --> 00:16:13,710
clustering FPGA board which is nice from M
boxes. For the particle ID then as I said

178
00:16:13,710 --> 00:16:18,780
earlier when we have to check off
detectors and they provide particle ID in

179
00:16:18,780 --> 00:16:23,100
different momentum ranges. For the first
chunk of detector, we replace everything.

180
00:16:23,100 --> 00:16:28,380
So, it was the gas you can see here a
sketch of the the optical readout system.

181
00:16:28,380 --> 00:16:32,730
So we replace the mirrors we replace the
quartz window, we replace the glass

182
00:16:32,730 --> 00:16:38,850
enclosure, and this is the focal plane
where you set the readout. So, the readout

183
00:16:39,150 --> 00:16:44,940
we've changed from HP DS in the in the
previous iteration abilities, and this

184
00:16:45,420 --> 00:16:51,300
will Deanna Fulton room for the multiply
chips from alma mater. And you can see

185
00:16:51,300 --> 00:16:58,260
here one of the fully assembled columns,
which has two types of one inch and a PMT

186
00:16:58,260 --> 00:17:05,430
s and then outside as the Larger two inch
painting for the color emitters and the

187
00:17:05,430 --> 00:17:09,540
neon system then we remove some detectors
that were only used for the level zero

188
00:17:09,540 --> 00:17:16,620
trigger and replace all of the readout
electronics. So the last thing I want to

189
00:17:16,620 --> 00:17:24,120
talk about is the upgrade to so this is a
new story that started two years ago,

190
00:17:24,210 --> 00:17:30,720
three years ago in 2017. In collaboration
with accelerator, so we want to increase

191
00:17:30,720 --> 00:17:37,020
the luminosity by a factor of 10. So we
run a 1.5 times 10 to the 34 and collect

192
00:17:37,020 --> 00:17:44,190
this integrated Lumosity 300%. Advance.
Now we have a much much harder environment

193
00:17:44,190 --> 00:17:48,600
to working on, we have to try to maintain
the same reconstruction performance as

194
00:17:48,600 --> 00:17:53,490
before. So we have to go to higher
granularity detectors and then we try to

195
00:17:53,490 --> 00:18:00,900
introduce timing information for the
tracking and for the particle ID So, this

196
00:18:00,900 --> 00:18:07,740
is just a sketch of the plan detectors we
have the tracking system, we will have new

197
00:18:07,740 --> 00:18:15,240
detectors. Already in the LS three, we
will install some stations which will be

198
00:18:15,240 --> 00:18:21,000
single layer bars for a DARPA si p EMS in
magnets. So just on the sides of the

199
00:18:21,030 --> 00:18:26,610
magnet to improve resolution on long
momentum tracks and the moon install a

200
00:18:26,640 --> 00:18:32,970
time of flight detector to give a decent
to improve the particle ID for low

201
00:18:32,970 --> 00:18:40,530
momentum tracks. And here this would be a
torch detector so it has caught quartz

202
00:18:41,490 --> 00:18:47,850
which is then read out with these micro
channel bmps outside to about two minutes.

203
00:18:48,090 --> 00:18:55,890
Okay, so this we have time resolution
around 15 Pico seconds per particle. Then

204
00:18:55,890 --> 00:19:00,870
we've completely replaced the tracking
stations again so we go in In the vertex

205
00:19:00,870 --> 00:19:08,100
detector go thinner sensors try doing the
timing. So the order of 200 picoseconds

206
00:19:08,100 --> 00:19:15,690
per hertz, and we get the order of 20
picoseconds per track in the tracking

207
00:19:15,720 --> 00:19:20,430
downstream of the magnet and we wouldn't
have the gain silicon detectors in the

208
00:19:20,430 --> 00:19:27,240
central region, assembling fibers outside.
And then for the for the particle ID there

209
00:19:27,240 --> 00:19:31,290
would actually be big changes in the
upgrade to so we would remove completely

210
00:19:31,470 --> 00:19:34,740
hadron calorimeter and replace it with
additional shielding for the Mian

211
00:19:34,740 --> 00:19:43,020
stations. And we would then install a new
electron electromagnetic calorimeter with

212
00:19:43,050 --> 00:19:47,730
much finer segmentation with the time than
planes so that we can have time resolution

213
00:19:47,730 --> 00:19:52,980
around 20 to 50 Pico seconds. And then
there's a lot of work going on on the

214
00:19:53,010 --> 00:19:59,760
ideas of using FPGAs and GPUs for the
online reconstruction. Van just wanted to

215
00:19:59,760 --> 00:20:04,410
flash Slide to say there's a so there's a
lot of r&d going on across the

216
00:20:04,410 --> 00:20:09,660
collaboration. It's we're looking for the
the color imagery, you can see an example

217
00:20:09,660 --> 00:20:15,720
here is tungsten with crystal fibers
inside. So, this has been used in a test

218
00:20:15,720 --> 00:20:22,710
beam, then the CMOS tracker, which will be
is proposed to be used in the inner region

219
00:20:22,710 --> 00:20:26,880
that's the tracker down stream with a
magnet we already started working on chip

220
00:20:26,880 --> 00:20:35,010
design. So, this will be based on the next
iteration of doubtless pixel v3 pixel. And

221
00:20:35,010 --> 00:20:37,890
then there are some development for new

222
00:20:41,910 --> 00:20:48,780
gas detectors for the service, you are
well protected for the means. And again,

223
00:20:48,810 --> 00:20:55,350
just something time in detector can be
used for the time of flight check off

224
00:20:55,350 --> 00:21:03,630
detector. So I come to my conclusions. So
an upgrade one bit started in 2008. And

225
00:21:03,630 --> 00:21:08,460
we're now coming to the end of the story
by the end of the construction phase. And

226
00:21:08,460 --> 00:21:14,010
the installation phase is well underway.
All of the detectors were in full

227
00:21:14,070 --> 00:21:18,570
production mode, after say the schedule
has become a lot more challenging than it

228
00:21:18,570 --> 00:21:26,160
was. And we have to continually assess the
impact of the pandemic. And finally, there

229
00:21:26,160 --> 00:21:32,970
was a lot of active r&d going on to
develop the baseline designs for the

230
00:21:33,030 --> 00:21:39,270
upgrade to and we expect to produce a
framework TDR where we would have first

231
00:21:39,270 --> 00:21:44,880
cost estimate and first proposed detector
at the end of next year, or in next year.

232
00:21:46,920 --> 00:21:47,310
Okay.

233
00:21:48,960 --> 00:21:53,880
Thank you very much. Mark. Are there
questions for Mark? If you have a

234
00:21:53,880 --> 00:21:58,530
question, please raise your hand and then
we'll see what's the participants

235
00:22:03,869 --> 00:22:07,949
So far, I don't see any questions. You
covered quite a lot of material in that

236
00:22:07,949 --> 00:22:17,789
talk. I guess 111 question I had are many
other talks mentioned the difference,

237
00:22:18,689 --> 00:22:23,279
let's say certain common electronics like
LPG, bt and so on. You're also making use

238
00:22:23,279 --> 00:22:24,629
of these, I presume as well.

239
00:22:24,960 --> 00:22:33,360
All of our readout is based on GBT. Sure.
Okay, all well controls the GBT SCA and

240
00:22:33,360 --> 00:22:37,170
then we have the GTX GTR x for the data
transmission.

241
00:22:39,450 --> 00:22:41,760
Okay, so the same as the others.

242
00:22:43,470 --> 00:22:52,890
Going for the last questions. Any, any
more questions coming up? Mark? I think we

243
00:22:52,890 --> 00:22:59,880
are at the end of the question. So thank
you very much, Mark. And thank you. I'd

244
00:22:59,880 --> 00:23:06,510
like Thank also with my co chair. Hello.
Yeah, we see you then, as well, the

245
00:23:06,510 --> 00:23:11,610
speakers. Yes, I think we'd like to thank
all the speakers for their picosecond

246
00:23:11,610 --> 00:23:19,080
timing, precision, and that we're finished
on time. Of course we'd like to thank the

247
00:23:19,080 --> 00:23:23,070
organizers as well. I don't know Paul, if
you wanted to say anything more.

248
00:23:24,089 --> 00:23:25,529
No, I think he said everything.

249
00:23:27,690 --> 00:23:32,130
And before we end I don't know if the
organizers want to say any last words

250
00:23:32,130 --> 00:23:34,140
before we close today's session.

251
00:23:35,190 --> 00:23:38,430
While we can just thank all the
participants for having attended today's

252
00:23:38,430 --> 00:23:44,220
session and contributed to the discussions
and we restart the game tomorrow at the