Move/Surf about sparta

[stanmoore1]

@oscillatingion not sure, will take a look

[stanmoore1]

@oscillatingion, I talked to @sjplimp and he pointed out this note from the doc page (see https://sparta.github.io/doc/move_surf.html):

After the surface has been moved, then all particles in grid cells that meet either of these criteria are deleted:

the grid cell is now inside a surface
the grid cell overlaps with a surface element that moved
This is to prevent particles from ending up inside surface objects.

So this fix is intended to adjust the surface and then allow for re-equilibration of the flow after a few timesteps. It currently won't "bounce" particles off the moving surface and impart momentum to the particles.

There may be a way to take some of the logic from #381 and apply it to the moving surface so only the particles that actually overlap with the surface are deleted, instead of all the particles in any grid cell that overlaps with a surface.

There may also be a way to couple a moving surface with the particles so they do bounce off the moving surface, but that would require modifications to the code.

Can you clarify your intended use case/what you need here?

[oscillatingion]

Thank you for the explanation. For my application, I will need to see the effect of the moving or spinning surface on the drag force. Thus, particles bouncing off the surface need to be counted. This type of application is like a bullet spinning during flight. Please advise me on how to couple the moving surface with particles. Best Regards, Charles /* . . . . . . . . . . . . . . . Charles C. Tseng, Ph.D. TMAS2 SME | CANVAS Inc. Advanced Missile Signature Center Arnold Engineering Development Complex Arnold AFB, TN 37389-9900 931-454-4178 . . . . . . . . . . . . . . . . . . . . . . . */ From: Stan Moore ***@***.***> Sent: Monday, February 27, 2023 12:25 PM To: sparta/sparta ***@***.***> Cc: TSENG, CHARLES C CTR USAF AFMC AEDC/718TS/TDS ***@***.***>; Mention ***@***.***> Subject: [URL Verdict: Neutral][Non-DoD Source] Re: [sparta/sparta] Move/Surf (Issue #399) @oscillatingion <https://github.com/oscillatingion> , I talked to @sjplimp <https://github.com/sjplimp> and he pointed out this note from the doc page (see https://sparta.github.io/doc/move_surf.html): After the surface has been moved, then all particles in grid cells that meet either of these criteria are deleted: the grid cell is now inside a surface the grid cell overlaps with a surface element that moved This is to prevent particles from ending up inside surface objects. So this fix is intended to adjust the surface and then allow for re-equilibration of the flow after a few timesteps. It currently won't "bounce" particles off the moving surface and impart momentum to the particles. There may be a way to take some of the logic from #381 <#381> and apply it to the moving surface so only the particles that actually overlap with the surface are deleted, instead of all the particles in any grid cell that overlaps with a surface. There may also be a way to couple a moving surface with the particles so they do bounce off the moving surface, but that would require modifications to the code. Can you clarify your intended use case and what you need here? — Reply to this email directly, view it on GitHub <#399 (comment)> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/A5U6Y7RPPEESK4GVYS2MH7DWZTWOFANCNFSM6AAAAAAVJVIMUM> . You are receiving this because you were mentioned. <https://github.com/notifications/beacon/A5U6Y7WXERAXWOEXRK25Q5LWZTWOFA5CNFSM6AAAAAAVJVIMUOWGG33NNVSW45C7OR4XAZNMJFZXG5LFINXW23LFNZ2KUY3PNVWWK3TUL5UWJTSWHUBQC.gif> Message ID: ***@***.*** ***@***.***> >

[sjplimp]

Charles - sorry, I'm still not clear on what you are asking for. a) Do you mean moving/spinning in the sense that a wall surface is sliding tangentially (e.g. for Couette flow) or a circle (in 2d) is spinning around its center point? If so, then the surf_collide command with its translate or rotate options can produce that effect. So that the post-collide velocities of particles which collide with the surface account for the effect of the translating/rotating surface. In this scenario, the surface itself (e.g. triangle) does not "move" in time. I.e. the coords of the 3 corner points of the triangle do not change. b) Do you mean you want a set of triangles which are like fan blades and you want to spin them? So that any instant the fan blades (triangles) are in different positions, and the particles collide with triangles that are moving normal to their plane? That is something SPARTA does not currently support. I'm also not clear it makes sense as a model in the DMSC context. Michael way want to comment on that. Steve

…

On Tue, Feb 28, 2023 at 10:27 AM oscillatingion ***@***.***> wrote: Thank you for the explanation. For my application, I will need to see the effect of the moving or spinning surface on the drag force. Thus, particles bouncing off the surface need to be counted. This type of application is like a bullet spinning during flight. Please advise me on how to couple the moving surface with particles. Best Regards, Charles /* . . . . . . . . . . . . . . . Charles C. Tseng, Ph.D. TMAS2 SME | CANVAS Inc. Advanced Missile Signature Center Arnold Engineering Development Complex Arnold AFB, TN 37389-9900 931-454-4178 . . . . . . . . . . . . . . . . . . . . . . . */ From: Stan Moore ***@***.***> Sent: Monday, February 27, 2023 12:25 PM To: sparta/sparta ***@***.***> Cc: TSENG, CHARLES C CTR USAF AFMC AEDC/718TS/TDS ***@***.***>; Mention ***@***.***> Subject: [URL Verdict: Neutral][Non-DoD Source] Re: [sparta/sparta] Move/Surf (Issue #399) @oscillatingion <https://github.com/oscillatingion> , I talked to @sjplimp <https://github.com/sjplimp> and he pointed out this note from the doc page (see https://sparta.github.io/doc/move_surf.html): After the surface has been moved, then all particles in grid cells that meet either of these criteria are deleted: the grid cell is now inside a surface the grid cell overlaps with a surface element that moved This is to prevent particles from ending up inside surface objects. So this fix is intended to adjust the surface and then allow for re-equilibration of the flow after a few timesteps. It currently won't "bounce" particles off the moving surface and impart momentum to the particles. There may be a way to take some of the logic from #381 < #381> and apply it to the moving surface so only the particles that actually overlap with the surface are deleted, instead of all the particles in any grid cell that overlaps with a surface. There may also be a way to couple a moving surface with the particles so they do bounce off the moving surface, but that would require modifications to the code. Can you clarify your intended use case and what you need here? — Reply to this email directly, view it on GitHub < #399 (comment)> , or unsubscribe < https://github.com/notifications/unsubscribe-auth/A5U6Y7RPPEESK4GVYS2MH7DWZTWOFANCNFSM6AAAAAAVJVIMUM> . You are receiving this because you were mentioned. < https://github.com/notifications/beacon/A5U6Y7WXERAXWOEXRK25Q5LWZTWOFA5CNFSM6AAAAAAVJVIMUOWGG33NNVSW45C7OR4XAZNMJFZXG5LFINXW23LFNZ2KUY3PNVWWK3TUL5UWJTSWHUBQC.gif> Message ID: ***@***.*** ***@***.***> > — Reply to this email directly, view it on GitHub <#399 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFA4LQT32CI6MMDBEY6VSU3WZYYPLANCNFSM6AAAAAAVJVIMUM> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[sjplimp]

@magalli - can you look at my comment/Q on this feature request ?

[oscillatingion]

Dear Steve, Thank you for the hint. My application is likely to be the Case A. I did not realize the angular velocity option could be found in the surf_ collide command instead of read_surf. Appreciate that! Best Regards, Charles /* . . . . . . . . . . . . . . . Charles C. Tseng, Ph.D. TMAS2 SME | CANVAS Inc. Advanced Missile Signature Center Arnold Engineering Development Complex Arnold AFB, TN 37389-9900 931-454-4178 . . . . . . . . . . . . . . . . . . . . . . . */ From: Steve Plimpton ***@***.***> Sent: Tuesday, February 28, 2023 1:30 PM To: sparta/sparta ***@***.***> Cc: TSENG, CHARLES C CTR USAF AFMC AEDC/718TS/TDS ***@***.***>; Mention ***@***.***> Subject: [URL Verdict: Neutral][Non-DoD Source] Re: [sparta/sparta] Move/Surf (Issue #399) Charles - sorry, I'm still not clear on what you are asking for. a) Do you mean moving/spinning in the sense that a wall surface is sliding tangentially (e.g. for Couette flow) of a circle (in 2d) is spinning around its center point? If so, then the surf_collide command with its translate or rotate options can produce that effect. So that the post-collide velocities of particles which collide with the surface account for the effect of the translating/rotating surface. In this scenario, the surface itself (e.g. triangle) does not "move" in time. I.e. the coords of the 3 corner points of the triangle do not change. b) Do you mean you want a set of triangles which are like fan blades and you want to spin them? So that any instant the fan blades (triangles) are in different positions, and the particles collide with triangles that are moving normal to their plane? That is something SPARTA does not currently support. I'm also not clear it makes sense as a model in the DMSC context. Michael way want to comment on that. Steve

On Tue, Feb 28, 2023 at 10:27 AM oscillatingion ***@***.*** <mailto:***@***.***> > wrote: Thank you for the explanation. For my application, I will need to see the effect of the moving or spinning surface on the drag force. Thus, particles bouncing off the surface need to be counted. This type of application is like a bullet spinning during flight. Please advise me on how to couple the moving surface with particles. Best Regards, Charles /* . . . . . . . . . . . . . . . Charles C. Tseng, Ph.D. TMAS2 SME | CANVAS Inc. Advanced Missile Signature Center Arnold Engineering Development Complex Arnold AFB, TN 37389-9900 931-454-4178 . . . . . . . . . . . . . . . . . . . . . . . */ From: Stan Moore ***@***.*** <mailto:***@***.***> > Sent: Monday, February 27, 2023 12:25 PM To: sparta/sparta ***@***.*** <mailto:***@***.***> > Cc: TSENG, CHARLES C CTR USAF AFMC AEDC/718TS/TDS ***@***.*** <mailto:***@***.***> >; Mention ***@***.*** <mailto:***@***.***> > Subject: [URL Verdict: Neutral][Non-DoD Source] Re: [sparta/sparta] Move/Surf (Issue #399) @oscillatingion <https://github.com/oscillatingion> , I talked to @sjplimp <https://github.com/sjplimp> and he pointed out this note from the doc page (see https://sparta.github.io/doc/move_surf.html): After the surface has been moved, then all particles in grid cells that meet either of these criteria are deleted: the grid cell is now inside a surface the grid cell overlaps with a surface element that moved This is to prevent particles from ending up inside surface objects. So this fix is intended to adjust the surface and then allow for re-equilibration of the flow after a few timesteps. It currently won't "bounce" particles off the moving surface and impart momentum to the particles. There may be a way to take some of the logic from #381 < #381> and apply it to the moving surface so only the particles that actually overlap with the surface are deleted, instead of all the particles in any grid cell that overlaps with a surface. There may also be a way to couple a moving surface with the particles so they do bounce off the moving surface, but that would require modifications to the code. Can you clarify your intended use case and what you need here? — Reply to this email directly, view it on GitHub < #399 (comment)> , or unsubscribe < https://github.com/notifications/unsubscribe-auth/A5U6Y7RPPEESK4GVYS2MH7DWZTWOFANCNFSM6AAAAAAVJVIMUM> . You are receiving this because you were mentioned. < https://github.com/notifications/beacon/A5U6Y7WXERAXWOEXRK25Q5LWZTWOFA5CNFSM6AAAAAAVJVIMUOWGG33NNVSW45C7OR4XAZNMJFZXG5LFINXW23LFNZ2KUY3PNVWWK3TUL5UWJTSWHUBQC.gif> Message ID: ***@***.*** <mailto:***@***.***> ***@***.*** <mailto:***@***.***> > > — Reply to this email directly, view it on GitHub <#399 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFA4LQT32CI6MMDBEY6VSU3WZYYPLANCNFSM6AAAAAAVJVIMUM> . You are receiving this because you were mentioned.Message ID: ***@***.*** <mailto:***@***.***> >

— Reply to this email directly, view it on GitHub <#399 (comment)> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/A5U6Y7UBFIGEQ7W6QUQAG4TWZZG43ANCNFSM6AAAAAAVJVIMUM> . You are receiving this because you were mentioned. <https://github.com/notifications/beacon/A5U6Y7X6TCQ3HR2RH7XSMNLWZZG43A5CNFSM6AAAAAAVJVIMUOWGG33NNVSW45C7OR4XAZNMJFZXG5LFINXW23LFNZ2KUY3PNVWWK3TUL5UWJTSWLIIMS.gif> Message ID: ***@***.*** ***@***.***> >

[stanmoore1]

@oscillatingion it sounds like your question was answered above, so I'm closing this issue. Let us know if you have other issues.