Choose timezone
Your profile timezone:
SPS BWS Task Force
→
Europe/Zurich
865/1-D17 (CERN)
BWS Taskforce meeting 02.06.2023
Present: Raymond Veness, William Andreazza, Benoit Salvant, Nabil El-Kassem, Christine Vollinger, Anthony Harrison, Thibaut Lefevre, Giannis Papazoglou, Heiko Damerau, Giulia Papotti, Michael Sullivan, Jonathan Emery, Kevin Shing Bruce Li, Ivan Karpov, Carlo Zannini, Federico Carra, Alexandre Mariet, Elena De La Fuente Garcia, +2 Italian Students.
https://indico.cern.ch/event/1292457/
Ray: We agreed on changing the orientation of the tank. If we don’t do a solution with one of the wire scanners, we should mention that this one is for failure analysis.
Christine: With the vertical one, if we install the feedthrough, it will be a little bit different. It would be good, if we could measure the beam spectrum.
Jonathan: Explanation of how the temperature was measured. The assumption is that the temperature uniform across the wire, which is probably not the case, therefore the local heating could be higher. The thermal coefficient for the resistivity was verified using a lab experiment. The temperature is going up, when we have beams. The temperature rise correlates with the increase in BCT, but not proportionally. 4 batches. With the negative coefficient, this high temperature is reasonable, because the temperature would diverge exponentially. The resistivity of the copper is ignored, because it’s super low and therefore negligible. Slide 7 it’s not flat top and we could acquire the bunch length. Slide 8: heating drops quickly once the beams are debunched (SFTPRO). 986 is the number of cycles. Slide 9: We’re having thermionic cooling due to release of electrons from the wire with the 4 batches beam. Difficult to say more than that the temperature threshold is reachced. Slide 10: The shortening of the bunch at the 4th is somewhat causing the sharp increase in temperature. However, need to cross-check as the bunch length is perhaps too long. The wires sublimate at temperatures above 2000K.
Benoit: We should check if the measurements are correct.
Elena: Power loss depends on Beam intensity, spectrum, impedance and coupling between resonance and beam power sprectrum. The shorter the bunch length the higher the power loss is. Slide 3: this power loss is happening periodically. The worst case is at 2.3 with a maximum dissipation of 7W on the wire with the ferrites installed, while the power dissipated on the ferrites will be 70W.
The outgassing does not start at a certain threshold, rather it begins gradually with every temperature increase.
Michael: Slide 3: the blue is without Ferrites. Introducing the Ferrites lowers Q and Impedance. Slide 5 shunt impedance multiplied by % loss on wire. Slide 6: The more ferrites we introduce the less the loss of energy on the wire is. The change in the frequency and dissipation on the wires is independent of the bunch length. Increasing the number of ferrites from 1 to 2 and 3 will not decrease the dissipated energy on the wire very much. Each increase in Ferrites will have a smaller effect, but the advantage is that the energy dissipation will be divided over three tiles instead of being concentrated on just one and the effect of coupling to more modes is not perhaps significant. Slide 9: Preliminary data have shown that installing a feedthrough would not cause a disadvantage in the power loss, resonance and dissipation on the wire. We will install ferrites on both WS and conduct a test by installing a feedthrough on only one of the WS. The feedthrough would provide us with additional information and data for future analysis. While the original coupler was found to have a minor negative impact, we are planning to improve the situation by changing the shape of the feedthrough, which is expected to be advantageous. Nevertheless, we need to agree on which WS we will install the coupler. H is the most important for operations. Not sure if coupler will make it better or worse. Agree to give a few more days for this discussion.
Anthony: The ferrites will probably outgas. We might have a problem with the vacuum group, and we’ll need ~6 hours more of pumping. It’ll be 36 hours instead of 30 hours. The pumps are not designed to remove more material that’ll be released from the ferrites so we might stumble into problems with the addition of 3 tiles. The pumping speed might be a problem.
Giannis: 35W uniformly distributed on the ferrite. The max temp is 380⁰c, but this is worst case scenario with steady state and max. power. The cooling through radiation is 99.5%, while conduction is only 0.5%. These simulations are done with a big ferrite. The simulations should be redone with the proper dimensions and position of the 3 ferrites.
Suggest adding throumo-strips to log temerature. Can we add other temperature logging?
Action Impedance: Provide Giannis with the proper dimensions of the ferrites and their positions.
Question again from vacuum on recovery time for vacuum.
Need a discussion with Mike Barnes for the recovery time impact on ABT.
William: Two spare WS will be ready by next week. The installation of the ferrites will be done in the tunnel. We have to all agree about the time needed for the intervention. One Vertical WS will be transformed into a horizontal WS.
In the next meeting we all need to discuss exactly what we’ll say to the committee IFC on Friday.
We all agree it would be good if we could install 3 Ferrite tiles.
Next meeting: Thursday 8th June at 9h00 in 864/2-B14.
There are minutes attached to this event.
Show them.
