[FFA_CEBAF_Collab] [EXTERNAL] Re: FFA corrector concept
Brooks, Stephen
sbrooks at bnl.gov
Tue Aug 23 13:09:50 EDT 2022
It *may* be possible to use fewer, but only after running error simulations. And typically, using fewer correctors means they have to be proportionately stronger each.
So it's better to think of a large number of channels but relatively low-power power supplies. Vendors can put several of these in a chassis in some cases. Digital control means there shouldn't be an enormous penalty for additional channels if the power stays the same, although we will have to run more cables.
-Stephen
________________________________________
From: Jay Benesch <benesch at jlab.org>
Sent: 23 August 2022 12:09
To: Chase Dubbe; Omar Garza; Joseph Meyers; Will Oren; Mike Bevins
Cc: ffa_cebaf_collab at jlab.org; Camille Ginsburg; Brooks, Stephen
Subject: Re: [FFA_CEBAF_Collab] [EXTERNAL] Re: FFA corrector concept
Colleauges,
In case it wasn't clear: Stephen's response below implies that every
window frame needs three power supplies, two for the dipoles and one for
the quad. Including splitters, perhaps 900 power supplies per arc.
Larger service buildings.
Jay
On 8/23/22 11:40, Brooks, Stephen wrote:
> If you have multiple energy beams in the same line, you need more correctors than for a single energy because you are correcting more "degrees of freedom". You can't wait for several cells and cancel an error downstream in a simplistic way because the beams undergo different phase advances. You'd have to use a complicated set of many correctors over several cells to do it.
>
> In particular, a dipole X and Y corrector on every magnet ensures all survey errors in the magnet centres can be cancelled out. These placement errors are amplified by the fact the gradients are fairly high in the FFA.
>
> In CBETA we had half this number: dipole X correctors on the F magnets and Y correctors on the D magnets and still could achieve reasonable, although not perfect, results with 4 beams. So, every magnet in CBETA had an iron frame but only 2 sides had copper coils.
>
> -Stephen
>
> ________________________________________
> From: Alex Bogacz <bogacz at jlab.org>
> Sent: 23 August 2022 10:59
> To: Brooks, Stephen
> Cc: Chase Dubbe; Omar Garza; ffa_cebaf_collab at jlab.org; Jay Benesch; Joseph Meyers; Will Oren; Mike Bevins; Camille Ginsburg
> Subject: Re: [FFA_CEBAF_Collab] [EXTERNAL] Re: FFA corrector concept
>
> Hi Stephen,
>
> That initial number came from me; a corrector every other cell….which you feel is too sparse. How about one corrector per cell? I think we need more thorough study, perhaps a MC simulation to quantify.
> We will further tackle it on Friday…Feel free to lead the discussion…
> Thanks.
>
> Cheers,
>
> Alex
> ___________________________________
> S. Alex Bogacz,
> Accelerator Physics Group Leader
> Center for Advanced Studies of Accelerators
> Jefferson Lab
> 12000 Jefferson Avenue,<x-apple-data-detectors://8>
> Newport News, VA 23606<x-apple-data-detectors://8>
> ___________________________________
> Sent from my iPhone
>
> On Aug 23, 2022, at 10:40 AM, Brooks, Stephen via FFA_CEBAF_Collab <ffa_cebaf_collab at jlab.org> wrote:
>
> 100 cells per 180 degree FFA arc would be 200 magnets, and 200 correctors. So for all FFA sections in the facility (two 180 degree arcs), 400 permanent magnets and 400 correctors.
>
> I think what Jay is saying makes sense too. Not sure where "50" came from.
>
> -Stephen
>
> ________________________________________
> From: FFA_CEBAF_Collab <ffa_cebaf_collab-bounces at jlab.org> on behalf of Jay Benesch via FFA_CEBAF_Collab <ffa_cebaf_collab at jlab.org>
> Sent: 22 August 2022 11:42
> To: Chase Dubbe; Omar Garza; ffa_cebaf_collab at jlab.org
> Cc: Joseph Meyers; Mike Bevins; Will Oren; Camille Ginsburg
> Subject: Re: [FFA_CEBAF_Collab] FFA corrector concept
>
> Chase,
>
> If there are only 50 BD and 50 BF per arc, then 100 assemblies per arc. Since the length of the BD+BF cell is about 125 cm, the arc circumference is ~250 m and the packing fraction is about 90%, I rounded up to 100 cells. I remember 88 and 90 cells per arc mentioned in weekly meetings. The splitters will need correctors as well and there are at least five cells in each of those. Since the splitters are individual lines by definition and there are six at each end of each arc in the present concept, add 2*6*5 to the number needed, for 150 total per arc. Alex's MMV.
>
> Jay
>
> ________________________________________
> From: Chase Dubbe <cdubbe at jlab.org>
> Sent: Monday, August 22, 2022 11:15 AM
> To: Omar Garza; Jay Benesch; ffa_cebaf_collab at jlab.org
> Cc: Will Oren; Camille Ginsburg; Mike Bevins; Joseph Meyers; Alex Bogacz
> Subject: Re: FFA corrector concept
>
> Good Morning Omar,
>
> To try to help, the FFA permanent magnets won't have a power requirement, but the Panofsky Quads that will be mounted along with them will. Based on the latest Magnet count sheet, there will be 50 of them per ARC (East/West) and reviewing Jay's technical write up, they should need 900W per magnet. He did mention that each would require two power supplies.
>
> With regards to the counts though, there is some discrepancy. Jay, for each of the ARCs, is the number 50 or 100? Looking at your paper, vs the magnet count sheet, you are specifying 100 per ARC vs. the 50. Also, do my assumptions on power and power supply requirements line up with your specification?
>
> [cid:e6d9ef9a-552c-471e-b389-b9c5801d6a03]
> *Snap shot from Jay's paper 'FFA Corrector Concept'
>
> Name L[cm] B[kG] G[kG/cm] Number of magnets
> BD 50 3.1 4.7 200
> BF 50 5.5 1.5 200
> Panofsky window quad, per Jay's sprec 100
> *Cells Taken from 'New Magnet and RF Inventory' Excel
>
> Thanks,
> Chase
> ________________________________
> From: Omar Garza <garza at jlab.org>
> Sent: Monday, August 22, 2022 10:51 AM
> To: Jay Benesch <benesch at jlab.org>; Chase Dubbe <cdubbe at jlab.org>; ffa_cebaf_collab at jlab.org <ffa_cebaf_collab at jlab.org>
> Cc: Will Oren <oren at jlab.org>; Camille Ginsburg <ginsburg at jlab.org>; Mike Bevins <mbevins at jlab.org>; Joseph Meyers <meyers at jlab.org>; Alex Bogacz <bogacz at jlab.org>
> Subject: Re: FFA corrector concept
>
> Good morning everyone.
> Does someone have the power requirements for the FFAs and new dipoles?
> Thank you.
> O'
>
> Omar Garza
> Deputy of Electrical Engineering Systems - Accelerator Reliability
> Thomas Jefferson National Accelerator Facility
> (757) 269-7257 garza at jlab.org<mailto:garza at jlab.org>
> ________________________________
> From: Jay Benesch <benesch at jlab.org>
> Sent: Monday, August 8, 2022 11:41 AM
> To: Chase Dubbe <cdubbe at jlab.org>; ffa_cebaf_collab at jlab.org <ffa_cebaf_collab at jlab.org>
> Cc: Will Oren <oren at jlab.org>; Omar Garza <garza at jlab.org>; Camille Ginsburg <ginsburg at jlab.org>; Mike Bevins <mbevins at jlab.org>; Joseph Meyers <meyers at jlab.org>; Alex Bogacz <bogacz at jlab.org>
> Subject: Re: FFA corrector concept
>
> Chase,
>
> It is my understanding that each of the permanent magnets will require
> one of these correctors. Only two of the three windings per assembly
> will likely be powered. Hence the estimate of 400 units and 800 power
> supplies. It seems to me better to build all the units with the three
> windings and power only two, to simplify fabrication and assembly. And
> we might need three in some locations. Rabbet joints assumed in the
> steel to simplify assembly.
>
> I'll send you the opc file for this under separate email.
>
> Jay
>
> On 8/8/22 11:27, Chase Dubbe wrote:
> Good Morning Jay,
>
> First, thank you for sharing all the other panofsky quad models. I have
> been out on vacation, so have not had a chance to look through them, but
> they are greatly appreciated.
>
> I am still trying to catch up, but for the cost estimate of these
> Picture Frame Correctors, is the intent to have one per each FODO set of
> permanent magnets in the FFA line, or will there be more or less? I just
> want to get a grasp at the quantity so that I can either use the cost
> estimate numbers Tommy Hiatt had developed, or perhaps go out for an RFI
> for cost.
>
> Thanks again for the help,
> Chase Dubbe'
> ------------------------------------------------------------------------
> *From:* Jay Benesch <benesch at jlab.org>
> *Sent:* Monday, August 8, 2022 10:18 AM
> *To:* ffa_cebaf_collab at jlab.org <ffa_cebaf_collab at jlab.org>
> *Cc:* Chase Dubbe <cdubbe at jlab.org>; Will Oren <oren at jlab.org>; Omar
> Garza <garza at jlab.org>; Camille Ginsburg <ginsburg at jlab.org>
> *Subject:* FFA corrector concept
> Attached please find a "picture frame" corrector concept which satisfies
> the requirements Stephen provided Friday. It should suffice for the cost
> estimate. Comments welcome.
>
>
> _______________________________________________
> FFA_CEBAF_Collab mailing list
> FFA_CEBAF_Collab at jlab.org
> https://mailman.jlab.org/mailman/listinfo/ffa_cebaf_collab
On 8/23/22 11:40, Brooks, Stephen wrote:
> If you have multiple energy beams in the same line, you need more correctors than for a single energy because you are correcting more "degrees of freedom". You can't wait for several cells and cancel an error downstream in a simplistic way because the beams undergo different phase advances. You'd have to use a complicated set of many correctors over several cells to do it.
>
> In particular, a dipole X and Y corrector on every magnet ensures all survey errors in the magnet centres can be cancelled out. These placement errors are amplified by the fact the gradients are fairly high in the FFA.
>
> In CBETA we had half this number: dipole X correctors on the F magnets and Y correctors on the D magnets and still could achieve reasonable, although not perfect, results with 4 beams. So, every magnet in CBETA had an iron frame but only 2 sides had copper coils.
>
> -Stephen
>
> ________________________________________
> From: Alex Bogacz <bogacz at jlab.org>
> Sent: 23 August 2022 10:59
> To: Brooks, Stephen
> Cc: Chase Dubbe; Omar Garza; ffa_cebaf_collab at jlab.org; Jay Benesch; Joseph Meyers; Will Oren; Mike Bevins; Camille Ginsburg
> Subject: Re: [FFA_CEBAF_Collab] [EXTERNAL] Re: FFA corrector concept
>
> Hi Stephen,
>
> That initial number came from me; a corrector every other cell….which you feel is too sparse. How about one corrector per cell? I think we need more thorough study, perhaps a MC simulation to quantify.
> We will further tackle it on Friday…Feel free to lead the discussion…
> Thanks.
>
> Cheers,
>
> Alex
> ___________________________________
> S. Alex Bogacz,
> Accelerator Physics Group Leader
> Center for Advanced Studies of Accelerators
> Jefferson Lab
> 12000 Jefferson Avenue,<x-apple-data-detectors://8>
> Newport News, VA 23606<x-apple-data-detectors://8>
> ___________________________________
> Sent from my iPhone
>
> On Aug 23, 2022, at 10:40 AM, Brooks, Stephen via FFA_CEBAF_Collab <ffa_cebaf_collab at jlab.org> wrote:
>
> 100 cells per 180 degree FFA arc would be 200 magnets, and 200 correctors. So for all FFA sections in the facility (two 180 degree arcs), 400 permanent magnets and 400 correctors.
>
> I think what Jay is saying makes sense too. Not sure where "50" came from.
>
> -Stephen
>
> ________________________________________
> From: FFA_CEBAF_Collab <ffa_cebaf_collab-bounces at jlab.org> on behalf of Jay Benesch via FFA_CEBAF_Collab <ffa_cebaf_collab at jlab.org>
> Sent: 22 August 2022 11:42
> To: Chase Dubbe; Omar Garza; ffa_cebaf_collab at jlab.org
> Cc: Joseph Meyers; Mike Bevins; Will Oren; Camille Ginsburg
> Subject: Re: [FFA_CEBAF_Collab] FFA corrector concept
>
> Chase,
>
> If there are only 50 BD and 50 BF per arc, then 100 assemblies per arc. Since the length of the BD+BF cell is about 125 cm, the arc circumference is ~250 m and the packing fraction is about 90%, I rounded up to 100 cells. I remember 88 and 90 cells per arc mentioned in weekly meetings. The splitters will need correctors as well and there are at least five cells in each of those. Since the splitters are individual lines by definition and there are six at each end of each arc in the present concept, add 2*6*5 to the number needed, for 150 total per arc. Alex's MMV.
>
> Jay
>
> ________________________________________
> From: Chase Dubbe <cdubbe at jlab.org>
> Sent: Monday, August 22, 2022 11:15 AM
> To: Omar Garza; Jay Benesch; ffa_cebaf_collab at jlab.org
> Cc: Will Oren; Camille Ginsburg; Mike Bevins; Joseph Meyers; Alex Bogacz
> Subject: Re: FFA corrector concept
>
> Good Morning Omar,
>
> To try to help, the FFA permanent magnets won't have a power requirement, but the Panofsky Quads that will be mounted along with them will. Based on the latest Magnet count sheet, there will be 50 of them per ARC (East/West) and reviewing Jay's technical write up, they should need 900W per magnet. He did mention that each would require two power supplies.
>
> With regards to the counts though, there is some discrepancy. Jay, for each of the ARCs, is the number 50 or 100? Looking at your paper, vs the magnet count sheet, you are specifying 100 per ARC vs. the 50. Also, do my assumptions on power and power supply requirements line up with your specification?
>
> [cid:e6d9ef9a-552c-471e-b389-b9c5801d6a03]
> *Snap shot from Jay's paper 'FFA Corrector Concept'
>
> Name L[cm] B[kG] G[kG/cm] Number of magnets
> BD 50 3.1 4.7 200
> BF 50 5.5 1.5 200
> Panofsky window quad, per Jay's sprec 100
> *Cells Taken from 'New Magnet and RF Inventory' Excel
>
> Thanks,
> Chase
> ________________________________
> From: Omar Garza <garza at jlab.org>
> Sent: Monday, August 22, 2022 10:51 AM
> To: Jay Benesch <benesch at jlab.org>; Chase Dubbe <cdubbe at jlab.org>; ffa_cebaf_collab at jlab.org <ffa_cebaf_collab at jlab.org>
> Cc: Will Oren <oren at jlab.org>; Camille Ginsburg <ginsburg at jlab.org>; Mike Bevins <mbevins at jlab.org>; Joseph Meyers <meyers at jlab.org>; Alex Bogacz <bogacz at jlab.org>
> Subject: Re: FFA corrector concept
>
> Good morning everyone.
> Does someone have the power requirements for the FFAs and new dipoles?
> Thank you.
> O'
>
> Omar Garza
> Deputy of Electrical Engineering Systems - Accelerator Reliability
> Thomas Jefferson National Accelerator Facility
> (757) 269-7257 garza at jlab.org<mailto:garza at jlab.org>
> ________________________________
> From: Jay Benesch <benesch at jlab.org>
> Sent: Monday, August 8, 2022 11:41 AM
> To: Chase Dubbe <cdubbe at jlab.org>; ffa_cebaf_collab at jlab.org <ffa_cebaf_collab at jlab.org>
> Cc: Will Oren <oren at jlab.org>; Omar Garza <garza at jlab.org>; Camille Ginsburg <ginsburg at jlab.org>; Mike Bevins <mbevins at jlab.org>; Joseph Meyers <meyers at jlab.org>; Alex Bogacz <bogacz at jlab.org>
> Subject: Re: FFA corrector concept
>
> Chase,
>
> It is my understanding that each of the permanent magnets will require
> one of these correctors. Only two of the three windings per assembly
> will likely be powered. Hence the estimate of 400 units and 800 power
> supplies. It seems to me better to build all the units with the three
> windings and power only two, to simplify fabrication and assembly. And
> we might need three in some locations. Rabbet joints assumed in the
> steel to simplify assembly.
>
> I'll send you the opc file for this under separate email.
>
> Jay
>
> On 8/8/22 11:27, Chase Dubbe wrote:
> Good Morning Jay,
>
> First, thank you for sharing all the other panofsky quad models. I have
> been out on vacation, so have not had a chance to look through them, but
> they are greatly appreciated.
>
> I am still trying to catch up, but for the cost estimate of these
> Picture Frame Correctors, is the intent to have one per each FODO set of
> permanent magnets in the FFA line, or will there be more or less? I just
> want to get a grasp at the quantity so that I can either use the cost
> estimate numbers Tommy Hiatt had developed, or perhaps go out for an RFI
> for cost.
>
> Thanks again for the help,
> Chase Dubbe'
> ------------------------------------------------------------------------
> *From:* Jay Benesch <benesch at jlab.org>
> *Sent:* Monday, August 8, 2022 10:18 AM
> *To:* ffa_cebaf_collab at jlab.org <ffa_cebaf_collab at jlab.org>
> *Cc:* Chase Dubbe <cdubbe at jlab.org>; Will Oren <oren at jlab.org>; Omar
> Garza <garza at jlab.org>; Camille Ginsburg <ginsburg at jlab.org>
> *Subject:* FFA corrector concept
> Attached please find a "picture frame" corrector concept which satisfies
> the requirements Stephen provided Friday. It should suffice for the cost
> estimate. Comments welcome.
>
>
> _______________________________________________
> FFA_CEBAF_Collab mailing list
> FFA_CEBAF_Collab at jlab.org
> https://mailman.jlab.org/mailman/listinfo/ffa_cebaf_collab
More information about the FFA_CEBAF_Collab
mailing list