Choose timezone
Your profile timezone:
FCC-ee Injector Design (CHART proposal) Coordination meeting 10
→
Europe/Zurich
Paolo Craievich
-
-
15:30
→
15:40
General information 10mSpeaker: Paolo Craievich
Bunch length at the linac end.
1 mm is too short to be injected in the collider. We have to find a solution to lengthen the bunch.
Specification for the collider is 4 mm and when we inject into the booster we cannot lengthen the bunch with present configuration.
Here from the chat:
16:04:08 From Frank Zimmermann to Everyone:
maybe one could foresee an arc with flexible R56 optics and then one could inject basically any bunch length into the booster, and optimize in simulations and also for the real machine
16:05:48 From Andrea Latina to Everyone:
hi Frank, in that case we would need to introduce an energy chirp which would require to run off-crest in the linac. we need to check the numbers..
16:08:42 From Katsunobu Oide to Everyone:
Not necessarily, depending on the R56 of the additional arc after the linac.
16:17:38 From Andrea Latina to Everyone:
True. We need to check how big the R56 would need to be, because I understand that the projected energy spread needs to be quite small..
16:21:18 From Barbara to Everyone:
Yes acceptable energy spread at 20 GeV in the Booster depends on DA and we are still working to improve it...
16:24:26 From Katsunobu Oide to Everyone:
Or, we may just inject the linac bunch to the booster, whose equilibrium bunch length is longer than the linac. This equivalently uses the BR ring as the bunch extender.
16:26:54 From Andrea Latina to Everyone:
This sounds like a very effective and pragmatic solution. Briefly, we need to understand what is the practical acceptance of the BR
16:29:38 From Barbara to Everyone:
I do not see the extension of the bunch length so far but could be that I have to check my bunch length evolution...
16:30:17 From Barbara to Everyone:
or one has to play with RF...
Follow-up discussion (Frank/Paolo):- Frank: yesterday I have been discussing with Catia Milardi. She seems to think that the use of the Damping Ring for the electrons may have been discarded. Earlier Katsunobu had argued that the damping ring could improve pulse-by-pulse stability But there is another argument in favour of using the damping ring for electrons. At the moment Barbara and Antoine plan to stay several seconds at top energy in the booster, prior to extraction. This time is mainly needed to damp the vertical emittance. If we inject a flat beam from the damping ring this time might be noticeably shortened. And it is much more energy- and time-efficient to damp in the damping ring (with a few bunches at low energy) than in the booster (with many bunches at high energy). So in my opinion we should try to retain the option of using the DR also for electrons. It costs almost nothing in terms of hardware, only some addt’l transfer lines. Even the same injection kicker could be used for both e- and e+. And the DR otherwise is not used when producing electrons. At the SLC, the DRs produced emittances down to 30 or 40 micron in x and 1 or 2 micron in y. Our damping rings a larger and could do better. Of course the alternative for us would be a flat beam electron gun, but I feel these flat guns are perhaps not so well established. (I might be wrong.)
- Paolo: When we discussed whether or not to use the DR for the electrons, we based it on the specifications we had at the injector end. These specifications were based on the specifications of the booster and the collider after a few meetings dedicated to optimising the interface parameters between the pre-injector and the booster. In my opinion, we should discuss these new proposals by organising a meeting dedicated to parameters as we did before. I will speak with Alexej about this meeting. I will try to comment on your points.
- DR for positrons: we got an update from Catia last week, but there is still a lot of work to be done to have robust dynamic acceptance. In addition, we have to design the positron energy compression and beam compression. Both tasks are in the Catia WP. In the past I used to organise meetings for this WP but in my opinion Catia should organise the work keeping in mind that in the end we will need to engineer the DR and transfer lines so that we can get a cost estimate for the review meeting next year.
- DR for electrons:
- improvement of pulse-by-pulse stability: If I understood Oide correctly, in SuperKEKB they have bunch-by-bunch stability problems in terms of intensity fluctuations and trajectory jitter. I think DR cannot improve intensity fluctuations and trajectory jitter will depend a lot on the dynamics in the linacs and photinjector. For this point we should try to write the specification on trajectory jitter at the linac end. We could also organise a meeting with the SuperKEKB colleagues to better understand the causes of their instabilities.
- damping the vertical emittance in the booster: For this point, too, we should try to write the specifications of the bunch at injection into the booster. So far we know that they need a bunch with an emittance of less than 50 um in both planes. Specification largely met by the bunch at the end of the linac at this time. If the vertical emittance specification should be of the order of 1 or 2 um at the pre-injector end then this would change the scenario completely and the pre-injector would have to be rethought, not only the damping ring but also the pre-injector and linacs.
- Definitely, we have to arrange a meeting to discuss the new specifications for the pre-injector.
-
15:40
→
16:00
Overall parameter optimisation 20m
-
Status of the High Energy Booster 10mSpeaker: Barbara Dalena (CEA-Irfu & Université Paris-Saclay (FR))
-
Strategy for Top-Up Injection 10mSpeaker: Rebecca Louise Ramjiawan (CERN)
-
-
16:00
→
16:20
WP3 - Positron source: target and capture system 20mSpeaker: Iryna Chaikovska (CNRS/IJCLab)
Configuration with a FC is actually in stand-by. Most likely the comparison studies won't be ready for the review meeting.
Hybrid scheme under study. CDR will be ready soon.
The SC technology is not a show stopper. The DPA/year is a limitation for the present AMD with this aperture.
Follow up (Anton/Paolo).
Anton:
I briefly wanted to come back to you concerning the operating conditions for the FCCee positron source (since it was mentioned in the meeting today). As you know, it is important for us to know the cumulative number of electrons on the target in order to estimate the long-term radiation damage in the target and coils. One of the key ingredients is the time for filling the collider from scratch. So far we assumed for Z-operation, that it takes 1035 s to fill the collider -> his number was derived by Salim [1]. We assume that during these 1035 s, the source is running *constantly* with the nominal drive beam parameters (1.8E10 electrons, 200 Hz). However, I saw in the slides for Barbara Dalena today (slide 2 in [2]) that filling the collider from scratch in Z operation takes 284 s. Did the numbers change so much, or do I misunderstand something? It would be important to make a table with the latest numbers (also for WW, ZH, ttbar) since this has a significant impact on conclusions/shielding design. For all beam modes, we would need:
*) Electron drive beam bunch charge (for filling collider from scratch and top-up injection)*) Number of electron bunches/pulse*) Pulse frequency*) Time for filling the collider, time for top up injection (and frequency of top-up injections)Paolo:
In Barbara D.'s slides, the values are for one species. To get the total filling times, you have to multiply by two. The filling times from scratch are 570 s using the SPS and about 500 s using the HE linac. The collider lifetime is 1085-1090 s, which means that the actual filling time meets this specification. We can meet this specs we need 2-bunch at 200 Hz with a charge of 5 nC for positrons and electrons. Based on the actual yield and capture of positrons in the DR, a drive beam with 1.8E10 electrons (2.88 nC) is a good assumption for your study considering a positron yield of about 2. Here Iryna can also comment. I cannot answer about the frequency of the top up injections but I will ask someone for this.
We had a dedicated meeting on this topic in March this year (https://indico.cern.ch/event/1133521/) where Salim presented the filling times for the two options, using the SPS or using the HE linac. Salim did not prepare a summary table, so the official table remains the one preset in September 2021 by Oide. However, the changes are small.
For now, those values still remains the reference for the pre-injector. In the next few days I will include this table in the excel file with the pre-injector specification parameters. Sometimes it is diffcult even for me to follow all the changes, but for the injector we will have to refer to that excel document on the collaborative workspace.Last piece of information: Oide et al. are doing studies to insert a polarisation wiggler in the collider that drastically reduces the lifetime. In this scenario, we would have to completely rethink the injector. However, the project coordinators told me that for now these are work in progress studies and that for now the specification table is as mentioned above.
Anton:Thanks for your reply. Indeed, in the indico link from Salim [1] which you provided I see that the injection time is half (i.e., around 570 sec) compared to Salim's previous estimate (1035 s) [2], which he had presented in a WP3 meeting. Unfortunately, we had missed this update.
Since the e+ and e- injections into the collider are interleaved, I suppose the e+ source does not run at full power during the 570 sec injection time but only during the e+ injections? Salim had given a Linac duty factor of 80.4% in [2]: does this reflect the fraction of 570 sec the e+ source runs at full power or does this refer to something else?
And indeed, for Z mode we assumed a drive beam with *one* bunch (1.8E10 electrons), with a frequency of 200 Hz (i.e. 1.8E10 *200 electrons impacting on the target per second). Does this change for WW, ZH, ttbar, or will the electron drive bunch charge/frequency be the same as for Z?
I think it would be highly beneficial, if we could agree on a table, listing the electrons on target per day for each mode, i.e., Z, WW, ZH, ttbar (maybe assumg 2x filling from scratch per day, while the rest is top-up injection). This would be very important for us since the long-term radiation damage depends on the number of e- on target.Paolo:The main reason for the change in filling times was the change in the number of buckets to be filled. Fewer buckets with more charge. I don't know what Salim's duty cycle refers to but what I can say is that using the HE linac we will have a train of 4800 rf pulses with 2 bunches (in total 9600 bunches). The trains of positrons and electrons last 24 s with a distance between trains of about 1 second or more (see attachment).The assumption of 1.8E10 electrons for the drive beam is an upper limit for us to keep until we have a more robust damping ring acceptance. Iryna can also comment this point.The other modes of operation of the collider involve fewer buckets to be filled and less charge. This means that the timing for filling and the filling pattern will be different but the positron source will have to provide less charge. At this time, we have not yet addressed the question of the other modes of operation.I will pass the question to Frank about the number of filling from scratch per day. I really have no idea. Same for the top-up operation.I understand exactly what you need to estimate the damage of the target in the medium and long term. -
16:20
→
16:40
WP4 - Damping ring and transfer lines 20mSpeaker: Catia Milardi (INFN e Laboratori Nazionali di Frascati (IT))
-
16:40
→
17:00
WP6 - e+ source in SwissFEL 20mSpeaker: Riccardo Zennaro
-
17:00
→
17:05
AOB 5m
-
15:30
→
15:40