Page 1 of 3 123 LastLast
Results 1 to 15 of 31

Thread: Hypervoxel Gas Leak with wind dynamics

  1. #1

    Hypervoxel Gas Leak with wind dynamics

    Hi all

    This is a the Gas Suspension system, Inside the duct, set of emitter follows the animation path of wind groups.
    Here the shape of the soft gas inside the duct has to controlled by hypervoxel gradient. You can see some gas leak outside the duct
    Also I have to match with the duct diameter. Sometimes the wind path is out of control. Why the animation path is not fitting in any perfect curve. Mostly irregular or infinitive curves. Is there any feature request for this? Or the Lightwave system designed this for purpose.People if you have solution please welcome.

    Thanks
    Thiyagu

    Click image for larger version. 

Name:	Query-GasLeak.PNG 
Views:	126 
Size:	726.7 KB 
ID:	146115Click image for larger version. 

Name:	Query-GasLeak.PNG 
Views:	126 
Size:	726.7 KB 
ID:	146115

  2. #2

    ParticleFX is far from perfect, if leaks occur i often use an Eraser object.

    also, this one could be solved using a proxy object i guess. (make the problem area more narrow)
    LW vidz   DPont donate   LightWiki   RHiggit   IKBooster   My vidz

  3. #3
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    15,118
    I donīt see why you would have particle leak and stray, unless you have more wind forces or collision objects affecting the main particles following the wind animation path, otherwise as erikals mentioned, you could try and use collision objects to eras the particles going off the path, but as I mentioned, a basic animation wind animation path default settings and a default particle emitter with default settings...it should just follow the path without any stray particles...so you must have more going on with particle settings or wind settings.

    Do you have the actual object set with collision? that may cause the issue, if you do not absolutely need that, then turn that off.

    And to clarify the nature of the leak, it isnīt hypervoxels leak, it is particle leak...a hypervoxel leak suggests the actual hypervoxels applied to the particles is the cause of the leak..and itīs not, hypervoxels is only the medium voxel sprite or volume to be used on to the particles.

  4. #4
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    6,509
    Quote Originally Posted by prometheus View Post
    And to clarify the nature of the leak, it isnīt hypervoxels leak, it is particle leak...a hypervoxel leak suggests the actual hypervoxels applied to the particles is the cause of the leak..and itīs not, hypervoxels is only the medium voxel sprite or volume to be used on to the particles.
    Actually, in this case, it may really be a "hypervoxels leak": As in, the particles themselves are confined within the pipe, but because hypervoxels are being used to draw volume medium around the particles, the hypervoxels may be set to draw to a larger radius than the particles themselves. That would allow the particles to be contained within the pipe, but the drawn hypervoxel volume medium to be seen outside the pipe (as seems to be occurring). Point is, if the hypervoxel radius is larger than the particle radius, then that means the particles need to be even further constrained within the pipe, otherwise the (larger) hypervoxel radius will cause the medium to appear outside the pipe even though the particle is within it.
    John W.
    LW2015.3UB/2019.1.4 on MacPro(12C/24T/10.13.6),32GB RAM, NV 980ti

  5. #5

    yep, could be that also.

     

    ...might also be fixed using "Particle Age"

    Last edited by erikals; 10-12-2019 at 04:40 PM.
    LW vidz   DPont donate   LightWiki   RHiggit   IKBooster   My vidz

  6. #6
    Electron wrangler jwiede's Avatar
    Join Date
    Aug 2007
    Location
    San Jose, CA
    Posts
    6,509
    Quote Originally Posted by erikals View Post
    yep, could be that also.
    I'm thinking that might be the problem because the "leaked" medium's boundary matches the pipe edge in such a "regular" manner. It doesn't look as though any particles are actually escaping the pipe, there'd be much more variation in the medium boundary drawn in that case -- instead, it looks to me as if the medium's just getting drawn slightly outside the edge of the pipe in that region (but still following it).

    Regardless, hope the OP can find the cause and solve it!
    Last edited by jwiede; 10-12-2019 at 04:39 PM.
    John W.
    LW2015.3UB/2019.1.4 on MacPro(12C/24T/10.13.6),32GB RAM, NV 980ti

  7. #7

    It doesn't look as though any particles are actually escaping the pipe,
    ah, true. Good catch!  

    (posted a possible Youtube video solution above)
    LW vidz   DPont donate   LightWiki   RHiggit   IKBooster   My vidz

  8. #8
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    15,118
    Quote Originally Posted by jwiede View Post
    Actually, in this case, it may really be a "hypervoxels leak": As in, the particles themselves are confined within the pipe, but because hypervoxels are being used to draw volume medium around the particles, the hypervoxels may be set to draw to a larger radius than the particles themselves. That would allow the particles to be contained within the pipe, but the drawn hypervoxel volume medium to be seen outside the pipe (as seems to be occurring). Point is, if the hypervoxel radius is larger than the particle radius, then that means the particles need to be even further constrained within the pipe, otherwise the (larger) hypervoxel radius will cause the medium to appear outside the pipe even though the particle is within it.
    Well...Initially I was aware of that possibility, but then reverted to particle leak since I thought the particles was only going through the semi transparent grey pipes..and in such case I would say it would be impossible to be a hv leak..
    but if the particles also flows through the brownish non transparent pipes...then truly as you say, the hv size may Indeed be the cause of it...and a second look at it..indeed that seem to be the case.

    Hvīs are not by themself a dynamic medium, so One need to ensure itīs radius isnīt too large as mentioned, or lower spread of particles so both the particle spread and radius doesnīt become larger than the inside part of the pipes.

    another way would be to use real fluids, from TurbulenceFD, or the gas solver..or run the sim in blender and send back to VDB for lightwave rendering.

  9. #9
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    15,118
    Quote Originally Posted by jwiede View Post
    I'm thinking that might be the problem because the "leaked" medium's boundary matches the pipe edge in such a "regular" manner. It doesn't look as though any particles are actually escaping the pipe, there'd be much more variation in the medium boundary drawn in that case -- instead, it looks to me as if the medium's just getting drawn slightly outside the edge of the pipe in that region (but still following it).

    Regardless, hope the OP can find the cause and solve it!
    I agree with you, he could use same hv particle size as to that of the actual emitters particle size, this however needs the pipes to act as collision objects so the radius of the particles do not go out of boundary, and that would ensure the hypervoxels not to go out of boundary as well..but the problem is that you will get very small hypervoxels and not much of hv textures to speak of for simulating smoke...that is why I would recommend using fluids, or fake the smoke with geometry and procedural texturing instead.

    Or...simply make the particles flow along itīs path in a non particle vibration and no spread along the path ..not untill it reaches itīs goal, all the particles should just be extremely tight flowing in a line, that way you can add a larger hv size that will not jump out based on particle position, but you will get a smoke flow look anyway.

  10. #10
    RETROGRADER prometheus's Avatar
    Join Date
    Aug 2003
    Location
    sweden stockholm
    Posts
    15,118
    blender smoke and two pipes, tpipe and elbowjoint, I didnīt even merge the two pipes, just placed it next to eachother, openGL render only, if you do have Lightwave 2018 Or 2019. you should be able to simulate it and save out cached VDB files and import the fluid sim to lighwave for rendering, if the wind animation path and hvīs still is troublesome.

    I guess you do not own turbulenceFD, so blender & Lightwave combo could be an alternativ, if you have 2019, vdb gas solver could also be of interest, but that takes some time and learn, as well as not as smooth to simulate with collision as in blender.

    exporting smoke sim from blender and Importing the vdb sequence to lightwave isnīt that hard, most troublesome may be how you want the smoke to flow in a more complex piping system.


  11. #11
    Quote Originally Posted by erikals View Post
    ParticleFX is far from perfect, if leaks occur i often use an Eraser object.

    also, this one could be solved using a proxy object i guess. (make the problem area more narrow)
    Hi
    There are two category 1.Materials 2.Gas
    I applied collider only for the MATERIAL to flow inside the pipes and cyclones which is necessary.Click image for larger version. 

Name:	particleonlycollider.PNG 
Views:	7 
Size:	64.4 KB 
ID:	146132

    But I think it doesnt need collider for the GAS flow because it is streamlined with the emitter to generate
    on Y axis only. The only way i needed to fix it by controlling the animation path of wind.
    Click image for larger version. 

Name:	GasParticleHighlight.PNG 
Views:	6 
Size:	104.7 KB 
ID:	146133

  12. #12
    Hi

    All your points are valid and I understand your views.
    No.its not particle leak. And some point I need to minimize using number of colldiers and simplify my workflow.The duct here is not a single piece of diameter. I had to complete the entire duct routing into a single continous. It can be controlled by Relative particle age.
    path following the different sizes or diameter.

    Click image for larger version. 

Name:	GasParticleHighlightHypervoxelOnly.PNG 
Views:	11 
Size:	344.0 KB 
ID:	146134

  13. #13
    Quote Originally Posted by jwiede View Post
    if the hypervoxel radius is larger than the particle radius, then that means the particles need to be even further constrained within the pipe, otherwise the (larger) hypervoxel radius will cause the medium to appear outside the pipe even though the particle is within it.
    That's right. The pipe or duct walls should be filled with gas, means if you do cross section it should reveal an uniform flow of gas in between the thickness of the metal. That close perfection is not possible due to change in sizes from start to end. But in Post production we can manipulate the exact thing by freezing the frame.

    Here, I have to spend some time for detailing the shape according to the duct sizes but some cases little tweaky.

    Click image for larger version. 

Name:	GasParticleHighlightHypervoxelLeak-.PNG 
Views:	10 
Size:	284.3 KB 
ID:	146135 Click image for larger version. 

Name:	GasParticleHighlightHypervoxelonlyVPR.PNG 
Views:	13 
Size:	94.3 KB 
ID:	146136

  14. #14
    Quote Originally Posted by erikals View Post
    yep, could be that also.

     

    ...might also be fixed using "Particle Age"


    Yeah, The entire gas duct size is modified with this Relative Particle Age only

    Click image for larger version. 

Name:	Duct Diameter variation.png 
Views:	4 
Size:	26.7 KB 
ID:	146137

    And the results are:

    Click image for larger version. 

Name:	GasParticleHighlightHypervoxelonlyVPR.PNG 
Views:	11 
Size:	94.3 KB 
ID:	146138

  15. #15
    Quote Originally Posted by jwiede View Post
    I'm thinking that might be the problem because the "leaked" medium's boundary matches the pipe edge in such a "regular" manner. It doesn't look as though any particles are actually escaping the pipe, there'd be much more variation in the medium boundary drawn in that case -- instead, it looks to me as if the medium's just getting drawn slightly outside the edge of the pipe in that region (but still following it).

    Regardless, hope the OP can find the cause and solve it!
    Click image for larger version. 

Name:	GasParticleHighlightYaxisonly.png 
Views:	8 
Size:	22.1 KB 
ID:	146142

    Hi,

    I made the particles lined up in a line of y axis to avoid escaping of particle and planned to control it over Hypervoxel gradient. So the size is just a line.

    Click image for larger version. 

Name:	GasParticleHighlight.PNG 
Views:	15 
Size:	104.7 KB 
ID:	146139
    Last edited by thiyaguthree; 10-14-2019 at 02:27 AM.

Page 1 of 3 123 LastLast

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •