Posted: 4/26/2013 7:11:33 AM EDT
| Can anyone recommend a package that works with supersonic flows? I have hyperworks, but it isn't accurate with supersonic flows. |
| EE here. I work with Ansys for computational EM problems so I use HFSS, Q3D, SIwave, etc. I don't mind them to much for my work and I understand that Ansys makes similar packages for CFD and thermo as well. (completely integrateable into their multiphysics environment I think). As has been said, you're going to pay an assload. It really isn't worth buying a one seat license for this stuff to design your own supressor. Your best bet is to get in with someone who already has it like a university. |
|
Quoted:
Add Flow3D to your list. I've worked a little with Ansys and Comsol, as well. I started down a similar road. Between time, effort and cost, it was easier to get an 07+SOT, machine the parts and test. I was just looking at Flow3D...looks like it could do the job. Looks pricey too though. I have access to hyperworks. I just wish it worked well with supersonic flows. A tad bit of a side topic, but to get an 07 + SOT, do you have to actively manufacture for commercial use or can you get one to build your own stuff anytime you want? |
|
Quoted:
A tad bit of a side topic, but to get an 07 + SOT, do you have to actively manufacture for commercial use or can you get one to build your own stuff anytime you want? You have to be in the business of dealing firearms to get the FFL. No minimums on manufacturing items. |
|
Quoted:
Fluent.  I would guess this as the answer. My area of specialty is structural mechanics, so it has been a long time since I did any CFD work, but way back when I worked with ANSYS I believe Fluent could probably do what you need. That said, what you are going to pay for licensing of the software would buy you many, many suppressors and some guns to host them. |
|
You're going to need one hell of a package to get good results for a simulation like that, and on top of that, you're going to need a lot background to understand whether your results are making any sense. Supersonic + subsonic mixed flow regimes will be pretty complicated.
Are you independently wealthy and specialized in an area of CFD and fluid flow analysis? If not, you're not going to get anything worthwhile out of an advanced simulation package, because it will require both. Use simplified models and methods where possible. Best of luck. |
|
Quoted:
You're going to need one hell of a package to get good results for a simulation like that, and on top of that, you're going to need a lot background to understand whether your results are making any sense. Supersonic + subsonic mixed flow regimes will be pretty complicated. Are you independently wealthy and specialized in an area of CFD and fluid flow analysis? If not, you're not going to get anything worthwhile out of an advanced simulation package, because it will require both. Use simplified models and methods where possible. Best of luck. I work with guys who are pro-gun and do CFD all day long that know a lot more about it than I do. I understand the basics as a mechanical engineer, but my focus is more on machine design and optics. As far as the cost of it...I was hoping I could stumble across some sort of student version and get access that way. At first glace, I've already found Fluent for $25 for a student seat. There's no way I'm paying tens of thousands for a seat to design a suppressor. I would rather do what was mentioned above (07+SOT) and just build and test, but I really don't want to sell anything. I hate laws sometimes. |
|
Quoted:
Quoted:
You're going to need one hell of a package to get good results for a simulation like that, and on top of that, you're going to need a lot background to understand whether your results are making any sense. Supersonic + subsonic mixed flow regimes will be pretty complicated. Are you independently wealthy and specialized in an area of CFD and fluid flow analysis? If not, you're not going to get anything worthwhile out of an advanced simulation package, because it will require both. Use simplified models and methods where possible. Best of luck. I work with guys who are pro-gun and do CFD all day long that know a lot more about it than I do. I understand the basics as a mechanical engineer, but my focus is more on machine design and optics. As far as the cost of it...I was hoping I could stumble across some sort of student version and get access that way. At first glace, I've already found Fluent for $25 for a student seat. There's no way I'm paying tens of thousands for a seat to design a suppressor. I would rather do what was mentioned above (07+SOT) and just build and test, but I really don't want to sell anything. I hate laws sometimes. Student version of Ansys has a 512k node/element limit. If thats the same for a Fluent mesh, that should be enough for an axisymmetric model. You're not going to be resolving boundary layers, and only need a rough resolution of the shocks. Don't know if the student version includes transient capability though. |
|
Quoted:
Quoted:
Quoted:
You're going to need one hell of a package to get good results for a simulation like that, and on top of that, you're going to need a lot background to understand whether your results are making any sense. Supersonic + subsonic mixed flow regimes will be pretty complicated. Are you independently wealthy and specialized in an area of CFD and fluid flow analysis? If not, you're not going to get anything worthwhile out of an advanced simulation package, because it will require both. Use simplified models and methods where possible. Best of luck. I work with guys who are pro-gun and do CFD all day long that know a lot more about it than I do. I understand the basics as a mechanical engineer, but my focus is more on machine design and optics. As far as the cost of it...I was hoping I could stumble across some sort of student version and get access that way. At first glace, I've already found Fluent for $25 for a student seat. There's no way I'm paying tens of thousands for a seat to design a suppressor. I would rather do what was mentioned above (07+SOT) and just build and test, but I really don't want to sell anything. I hate laws sometimes. Student version of Ansys has a 512k node/element limit. If thats the same for a Fluent mesh, that should be enough for an axisymmetric model. You're not going to be resolving boundary layers, and only need a rough resolution of the shocks. Don't know if the student version includes transient capability though. Hmm. Good to know. I was looking at building a monocore type design which wouldn't be symmetric about the axis. I'll have to look into the student version of fluent more to see what its limitations are. Has anyone used its acoustic modeling? At first glance it seems pretty cool and might be able to give me an estimate on sound suppression. Am I wrong in its capability there? |
|
Quoted:
Has anyone used its acoustic modeling? At first glance it seems pretty cool and might be able to give me an estimate on sound suppression. Am I wrong in its capability there? Heh, I'm structural (Ansys, Lsdyna, Dyrobes, Nastran, etc). Know a little about Fluent from co-workers. Acoustics is PFM as far as I'm concerned. Good luck! |