Comparision of down time for 2 runs 86 hours total 54 hours total RF down time distribution Summary Operation experience Cool down and warm up Gun CC1 and CC2 CM2 Future improvement Cool down and warm up for CC1CC2 CM ID: 1014112
Download Presentation The PPT/PDF document "RF operation experience and future impro..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
1. RF operation experience and future improvement
2. Comparision of down time for 2 runs86 hours total54 hours total
3. RF down time distribution
4. SummaryOperation experienceCool down and warm upGunCC1 and CC2CM-2Future improvement
5. Cool down and warm up for CC1,CC2, CMThings to do before cooling downwarm coupler conditioningKlystron and modulator(1 week)Things to do after cooling downTune cavity to resonance using LLRF (2 days)Conditioning with LLRF (2-5days)Things to do before warm upTune cavity to off resonance
6. Gun operationsOverall gun operation is OK2 long shutdowns due to the modulator issue (trip on insufficient undershoot voltage) HLRF people eventually was able to identify the cause of the problem after happening multiple times. It’s fixed now.Gun gradient and phase related to laser is fairly stable. Phase scan everyday shows pretty much the same shape.One thing missed from downtime log is that every time gun is down it will take at least 40 minutes for beam to be back because of water temp swing and gradient ramp up, making cave access much more difficult.
7. CC1 and CC2 operationOverall it is OK as well. Phase is relatively stable.Both cavity tripped multiple times.CC1 is normally on PD RF and Mod RF (at least 5 times)CC2 is normally due EF RF (at least 10 times)Most of the time operator can take care without consulting expert, which is the reason why the downtime is very little due to CC1 and CC2. We dtill need to consult with Elvin to solve this for next runIt did happen once that due to the cryo temp change cavity is detuned and need to change the cavity tuner motor to bring back the resonanceLater in the run CC2 gradient dropped from 14MeV to less than 13MeV. Cause still need to be identified, but is likely the couplerMysterious 855KHz from CC2 LLRF feedback
8. 858KHz from CC2/ figure from RandyAttention need be paid to start the next run. Need to contact CD(Ted Zmulda/Gustav Cancelo) for possible fix if possible.Magnitude vs different CC2 feedback gain
9. CM-2 operationNo downtime logged for CM because it’s used less frequently, only for 300MeV and few other shifts. A lot of time was spent to figure out how the cavity phase tuner works. Cavity 6 is limiting the gradient we could achieve, around (247MeV). Not too much thing we could do at this time. We used Lorentz force detuning feedback all the time. However LLRF feedback is not completely optimized yet. It remains to see whether we will need it for IOTA operation. (unlikely)
10. Future improvement:GunModulator issue is identified so we will hope that means much less down time from gun next run.Gun temp problemUsing auralee’s machine learning algorithum to control the water temperature to reduce the water temp swing problem.Maybe using PLC to do some feed forward control to decrease the swing as well.Long bunch train studyIn principle at least for 1Hz operation we should be able to run 1ms, although cavity temp set will be different for 1ms operation and 300us operation. If we need to do 5Hz Ding need to be consulted before.
11. Future improvement: CC1 and CC2 We will need CC1 and CC2 to be stable. Need to figure out why CC2 gradient is degrading. Most likely the degradation is from couplerOne of the ways might be to apply bias voltage to the coupler antenna. (Other labs also do this according to Elvin.)Longer conditioning is not going to help in this case. What we will need is to do high power conditioning. One feasible way is to switch the CM klystron to power up CC1 and cc2 to do the conditioning. This is significant work requiring modifying the waveguide and re-orienting the interlock scheme for both CM-2 and CC’s. Something like this was done in reverse for long pulse tests several years ago. Long bunch operation: with 13MV/m, long pulse operation is probably OK.
12. Future improvement: CM-2Better start up sequencing to start the cryomoduleBecause of the limitation of cavity 6 we may want to consider the installation of the waveguide distribution system with motorized VTO’s which will allow us to change the power distribution outside the cave. That way we can put cavity 6 on resonance but in lower power while powering other cavities higher. Two such distribution systems are onsite. It may take 6 weeks to install it then test – this requires removal of the existing one and installation with the roof off. Discussion with John Reid is needed.Long bunch train and high energy operation may benefit by this move.