Minutes of the LRFF Task Force
13th meeting on Tuesday 11/09/2012 (09:00-11:00 max, 6-R-018)
LRFF members: Alessandro Bertarelli (AlessandroB), Alexej Grudiev (AG), Benoit Salvant (BS), Elias Metral (EM), Fritz Caspers (FC), Giuseppe Bregliozzi (GB), Hugo Alistair Day (HD), Jose Miguel Jimenez (JMJ), Marco Garlasche (MG), Mike Barnes (MB), Olav Ejner Berrig (OB), Oleksiy Kononenko (OK), Oliver Aberle (OA), Ralph Assmann (RA), Raymond Veness (RV), Rhodri Jones (RJ), Roberto Losito (RL), Sergio Calatroni (SC), Stefano Redaelli (SR), Vincent Baglin (VB), Vittorio Parma (VP), Wim Weterings (WW).
Present/Excused: AlessandroB, AG, BS, EM, FC, GB, HD, JMJ, MG, MB, OB, OK, OA, RA, RV, RJ, RL, SC, SR, VB, VP, WW, ChristineV.
1) Comments on the last minutes + Actions
- Discussion about the ferrite for BSRT => See talk later, with slides from MG.
2) Review of ferrites used at CERN to damp HOMs: references, manufacturers, thermal treatments, etc. (all the groups: TE/VSC, EN/STI, TE/ABT, BE/RF, BE/BI, BE/ABP, EN/MME, etc.):
- EN/MME (review of ferrites on collimators): Slides already available since 31/07/12 (but we had no time to discuss them yet).
- New TCTP with TTT2-111R (reminder: R means antistatic). These are tiles from Skyworks.
- Waiting for results from vacuum people for the outgassing.
- FC used this type of ferrite in the AAA 25 years ago and it was very good vacuum wise.
- TCTVB (& TCLIA) => 4S60 (NiZn tiles from ferroxcube). Hybrid solution with both RF fingers and ferrite. In point 2 were already removed and they will be removed in 8 during LS1. There is also ferrite in the inside and outside inlets.
- Reminder from FC to stop using 4S60 and use TT2-111R:
- => 2 reasons: Low curie point and pb with manufacturer with brittleness => Liberate dust and we saw cracks in it.
- Reminder from SC: we should try and avoid anything with Zn.
- Ferroxcube might have also the equivalent of TT2-111R (4C65? tbc). FC is discussing with them.
- Question from me: why did we use 4S60 and 8C11 in the fast if the TTT2-111R was known to be better since a long time? FC mentioned that the 8C11 is a good ferrite for kicker but we look here for a bad ferrite. The losses are not excellent at least until ~ 500 MHz and we are aiming for high loss in the ferrite.
- FC's sentence of the day: "data sheet should not contain shit data".
- AlessandroB mentioned that there is always a missing information, the thermal emissivity, which is very important for radiation. So far they used 0.8. General belief is that it is black, i.e. between 0.8 and 1, but it is difficult to know...
- The power loss in this ferrite was calculated and should be confirmed.
- Discussion with FC about the use of both ferrite vs. RF fingers (which we had already in the past and it was decided to go with this scheme).
- TE/ABT: Slides already available since 05/06/12 (but we had no time to discuss them yet).
- Damping ferrites are used only in the MKI tank.
- TEM mode damping ferrite.
- Type of ferrite (from ferroxcube): 4M2 (NiZn) and 4B3 (NiZn). FC decided to use them because there were low frequency and were easily manufactured.
- Note that a ceramic capacitor with carbon deposit resistors is used for the SPS septa and protection elements instead of ferrite, as suggested in the past by FC.
- No issue seen until now.
- Why treating this at 1000 deg? We should review this with VB. Maybe just because of convenience (for the volume also). We should have some guidelines somewhere.
- BE/BI: Slides already available since some time (but we had no time to discuss them yet).
- Discussion with FC in the past.
- Detailed information could not be provided. Action 1 (BI): Can we find more information?
- Ferrite is used also for the PS WS (ferroxcube 4SD60), SPS BWS (ferroxcube 4S60, same as before?), PSB and LHC wire scanners (either 4S60 or Philips ferrite), PS WCMS (8C11)...
- FC mentioned that for the BSRT, he was contacted in the past and then there was no follow-up and finally the 8C11 ferrite was chosen, but he did not give his approval => We should be very careful in the future to follow all the design procedure and one should make sure that anyone comes at an early stage such that we can still act in case it is needed.
- Reminder from RJ: We don't have ferrite in the Schottky. TomK calculated in the past the power to be lost and the cooling of the ferrite could not be guaranteed => It was decided with FC to let it like this, i.e. without ferrite.
- AlessandroB mentioned that the mu' and mu'' depend on temperature, which is true. We don't take this into account in our impedance estimates => To be simulated (Action 2 below).
- At one of the next meetings, ChristineV will present some of here recent measurements (with FC) to characterize the TT2-111R ferrite's properties => Planned for the moment for the meeting on 02/10/2012.
3) Discussion about the ferrite for the BSRT => Slides from MG (Numerical evaluation of RF induced heating): pptx
- MG used some FEM (Finite Element Modeling) with an axisymmetric case to have faster simulations (even if it is not axisymmetric in reality). Furthermore, the mirror was not simulated.
- From BS, the power loss in the ferrite should be between 10 W and 50 W depending on the mirror (and its coating). FC and ChristineV think from their measurements that there is a resistive layer.
- The mirror is not considered for the emissivity as it is not seen by the ferrite. For the ferrite they used 0.8 as it is almost a black body. It is reasonable to start with this value and it could be 0.9 but we expect a 10% effect => To be checked.
- Case 1 studied: P = 50 W, the ferrite radiates to ambient (blackbody of emissivity= of 1), the other components are neglected, the ambient temparture is 22°C.
=> For the best case scenario for evacuation only through radiation, 390 deg are found in the ferrite and it cannot be better than this if only radiation is taken into account. The simulations are in agreement with computations from AlessandroB (who for 10 W finds already ~ 200 W). Reminder from FC: we could have tunneling effect as the distance is ~ 2-3 microns and we are discussing wavelengths of ~ 10 microns or so => Should be studied in detail and could help. Reminder: ferrite is usually only touching at few (3) points and there is almost no conduction.
- Case 2 studied: P = 50 W, the ferrite radiates to other components, no contact (all inner components exchange heat only through radiation, also with external vacuum tank), vacuum tank cooled by external free convection (α = 5 W / m2 K, TAMBIENT = 22°C).
=> 560 deg are found.
- Case 3a studied: P = 50 W, the ferrite radiates to other components, no contact (all inner components exchange heat only through radiation, also with external vacuum tank), vacuum tank at constant temperature of 22°C.
=> 557 deg are found => The cooling of the external tank is not interesting.
- Case 3b studied: P = 50 W, the ferrite radiates to other components, contact between mirror support and flange (all other inner components exchange heat only through radiation, also with external vacuum tank), vacuum tank at constant temperature of 22°C.
=> 553 deg are found => The cooling method is not so interesting.
- Case 4 studied: P = 50 W, the ferrite radiates to other components, all other inner components exchange heat only through radiation, also with external vacuum tank), conduction (thermal contact chain from ferrite to flange), vacuum tank cooled by external free convection (α = 5 W / m2 K, TAMBIENT = 22°C).
=> 460 deg are found => How the tank is cooled becomes important and ~ 100 deg can be gained.
- Case 5 studied: P = 10 W, the ferrite radiates to other components, no contact (all inner components exchange heat only through radiation, also with external vacuum tank), vacuum tank cooled by external free convection (α = 5 W / m2 K, TAMBIENT = 22°C).
=> 286 deg are found.
- Preliminary conclusions:
- For P = 50 W, if heat is evacuated only through radiation it is impossible to keep ferrite temperature below Tcurie of the grade (case 2-3).
- Cooling also by conduction is highly beneficial (case 4).
- Changing ferrite to a grade with higher Tcurie is effective only if RF induced power is 10W condition (case 5).
- FC asked if they could provide a new ferrite with the highest Curie temperature => FC will discuss with Trans-Tech. But then we have the issue with other components which might not withstand this temperature.
- Reminder from FC: the power loss would be lower if the ferrite could lose more (which could seem counter-intuitive).
- What would happen without ferrite (as in all the previous cases it would mean that the ferrite is operating above the Tcurie and therefore it should not be effective)?
- All indications seem to be consistent with numbers found here and therefore we should have temperatures much above Tcurie. But if above we should lose the ferrite properties then we should lose more than 10 times more and then (Reminder: the point where the ferrite loses its properties is quite sharp in temperature for the ferrite kickers within few degrees). We do have a strong RF loss mechanism in the mirror itself. Could it be that when the ferrite does not work we are lucky that we do not hit the resonance anymore?
- Mirror = silicon + dielectric coating and it seems there is a resistive layer. Can we exclude that there is a 10 nm metal layer?
- The coating of the mirror is made to have a better reflection, it was an optical choice. Al coating is standard but then it leads to the worst power loss (50 W).
- Conclusion: either we overestimate the power loss or we operate above Tcurie => We have to replace the CuBe by inconel and ferrite by TT2-111R. The next decision to take is about the mirror itself.
4) Actions to be taken for the next meeting
- Old actions.
- New actions:
- Action 1 (BI): Can we find more information about the damping ferrite used in the LHC?
- Action 2 (Impedance team): Take into account the temperature dependence of the mu' and mu" of the ferrite for the power loss estimate.
5) Miscellaneous
- The next (14th) meeting will take place on 25/09/2012 between 09:00 and 11:00 (max.) in room 6-R-018 => Agenda:
1) Intermediate status report of this task force (EM).
- See preliminary agendas for the next meetings.
Minutes by E. Metral, 24/09/2012.