Hi,
as a Wood Engineering student i'd like to use Comsol to simulate the behaviour of enclosure materials for the design of loudspeaker.
I'm familiar with some of their application notes but i can't create a good model for what i have in mind, which is in my opinion relatively easy.
This is what i'd like to be able to compute:
1.
INPUT:
the TS parameters, the enclosure design (i mean closed or vented with defined alignment) and the input voltage
OUTPUT:
the theoretical maximal force in Newton that the driver can develop.
i know that part of it would be damped and another part transformed in acoustic waves. i would just go with the worst case scenario, and the model would be simpler.
(this part could be easily done with Leap btw)
2.
INPUT:
the Force in Newton computed in 1. , an actual enclosure designed in Solidworks and synchronized trough livelink.
i would assign that Force to the flange where the driver would be attached.
OUTPUT
the mechanical stresses and the acoustic radiation of the enclosure due to those stresses.
Just as info; i'd like to create this model to simulate different wood composite materials that i'll have to research.
I wanna keep this model SIMPLE, because i can't spend months to thoroughly learn this software, that is amazing btw.
I'll appreciate any help and of course i'll be glad to share my research with you, if interested.
as a Wood Engineering student i'd like to use Comsol to simulate the behaviour of enclosure materials for the design of loudspeaker.
I'm familiar with some of their application notes but i can't create a good model for what i have in mind, which is in my opinion relatively easy.
This is what i'd like to be able to compute:
1.
INPUT:
the TS parameters, the enclosure design (i mean closed or vented with defined alignment) and the input voltage
OUTPUT:
the theoretical maximal force in Newton that the driver can develop.
i know that part of it would be damped and another part transformed in acoustic waves. i would just go with the worst case scenario, and the model would be simpler.
(this part could be easily done with Leap btw)
2.
INPUT:
the Force in Newton computed in 1. , an actual enclosure designed in Solidworks and synchronized trough livelink.
i would assign that Force to the flange where the driver would be attached.
OUTPUT
the mechanical stresses and the acoustic radiation of the enclosure due to those stresses.
Just as info; i'd like to create this model to simulate different wood composite materials that i'll have to research.
I wanna keep this model SIMPLE, because i can't spend months to thoroughly learn this software, that is amazing btw.
I'll appreciate any help and of course i'll be glad to share my research with you, if interested.
Honestly, if you don't have the time to climb the awful learning curve, then COMSOL is probably not for you. 
Anyhow, there is at least one tutorial model of a loudspeaker that you might be able to poach most of the necessary details you need from it. Then, yes, you'll need to patch that into your enclosure model.
That said GIGO is the name of the day with any of these simulation packages so make sure you have a *good* idea about your wood models (plywood anisotropy, empirical models about joints, etc).
I'd be making sure that I've squeezed every last drop out of the FEA built into Solidworks before I swapped over to COMSOL. Sorry to be the bearer of bad news.
If the goal is to learn COMSOL and the results are essentially unimportant, then absolutely go for it! Best time to learn the software for future work is right now.
Anyhow, there is at least one tutorial model of a loudspeaker that you might be able to poach most of the necessary details you need from it. Then, yes, you'll need to patch that into your enclosure model.
That said GIGO is the name of the day with any of these simulation packages so make sure you have a *good* idea about your wood models (plywood anisotropy, empirical models about joints, etc).
I'd be making sure that I've squeezed every last drop out of the FEA built into Solidworks before I swapped over to COMSOL. Sorry to be the bearer of bad news.
If the goal is to learn COMSOL and the results are essentially unimportant, then absolutely go for it! Best time to learn the software for future work is right now.
Hi,
thanks a lot for the honesty, man.
Well maybe i wasn't clear enough, i'm sorry for that.
I DO wanna learn Comsol, i think it will be a very good tool and to be honest i think it's fun, most of the times.
I'm looking for seminars and courses, they are expensive but i guess worth it.
I think you're talking about the application note "loudspeaker driver in a vented enclosure"; in fact it's a great example but it's really difficult for me to modify it for my needs.
as you said it's not trivial to model wood based materials, but that is what i'm studying and where i'd like to put my effort. i'm sure you understand what i mean.
you're not bearer of bad news, you've been honest and that is something i do appreciate.
thanks a lot for the honesty, man.
Well maybe i wasn't clear enough, i'm sorry for that.
I DO wanna learn Comsol, i think it will be a very good tool and to be honest i think it's fun, most of the times.
I'm looking for seminars and courses, they are expensive but i guess worth it.
I think you're talking about the application note "loudspeaker driver in a vented enclosure"; in fact it's a great example but it's really difficult for me to modify it for my needs.
as you said it's not trivial to model wood based materials, but that is what i'm studying and where i'd like to put my effort. i'm sure you understand what i mean.
you're not bearer of bad news, you've been honest and that is something i do appreciate.
I haven't gone to any of the seminars/etc, and I'm sure that'd speed things up, but learned a ton from looking at their example files and working from there (like you have), so it's a matter of starting from there and then building up more and more sophisticated models.
But, yes, I'd try to *remove* as much of that model as possible (or one of the BEM examples) and get used/an intuitive feel to what you're modifying from the file (size/forces/etc) and then you can take those core principles and combine it with vibration/structural modes from other tutorials.
The nice thing is you can keep breaking and breaking these files and learn a lot really quick. Hopefully you can simplify the mesh to something you can simulate in short time without convergence issues.
Seriously, good luck!
But, yes, I'd try to *remove* as much of that model as possible (or one of the BEM examples) and get used/an intuitive feel to what you're modifying from the file (size/forces/etc) and then you can take those core principles and combine it with vibration/structural modes from other tutorials.
The nice thing is you can keep breaking and breaking these files and learn a lot really quick. Hopefully you can simplify the mesh to something you can simulate in short time without convergence issues.
Seriously, good luck!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.