Discussion Forum

September 2, 2010 4:15am UTC

saw device comsol 4.0

Hi.. I’m a beginner for this COMSOL application. As a start, I had modeled a SAW device using lithium niobate at 218 Mhz in order to find the corresponding eigenfrequency and frequency response. I did it using the earlier version of COMSOL which is COMSOL 3.5 and had obtained the correct result. However, when I tried to model and simulate it using the newer version which is COMSOL 4.0, the result obtained is not the same as the previous one. Attached is my model for your kind reference.



1) Can you explain to me what exactly went wrong with my model? 2) Can you assist me to find the eigenfrequency and freq response of the model attached using the newest version of COMSOL 4.0

Warm regards,

ANNA

Reply  |  Reply with Quote  |  Send private message  |  Report Abuse

February 5, 2011 3:09pm UTC in response to Anand Arcot Narasimulu

Re: saw device comsol 4.0

Hi

I have only 4.1 but when I load in the file, I see:

1) reasonnable geometry that I understand ;)
2) some material constants, but that are not in use as they are not referring to any domain
3) a PZT physics, one for all materials that are redefining all domains as PZT material, which I believe is not the desired case, or ?
4) an eigenfrequency analysis around 218 MHz

You should add a linear exansion node for the Aluminium, and either decide if you use the material data or you define the material data as User specified in the physics node (but not both)

Since you solve also for V, you need to ensure that you have enough BC's for (u,v,w) the structural part (fixed i ok), as well as for V the voltage (you need a GND and a V source). You should also check what impedance you have on the electric circuit as a PZT has different stiffness if the electrodes are open circuit or shorted/grounded

--
Good luck
Ivar

Reply  |  Reply with Quote  |  Send private message  |  Report Abuse

March 7, 2011 12:46pm UTC in response to Ivar Kjelberg

Re: saw device comsol 4.0

Hello Sir,
I have plotted the current at the electrode where it was 1V and for calculating "I" i have used (pzd.nJ)* 1e-6[m]. I am here with attaching my word file with images for better understating of my model (the way i have set my boundary conditions). please correct me so that i can calculate the resonance frequency. Thanks in Advance.

Thanks
Anand

Reply  |  Reply with Quote  |  Send private message  |  Report Abuse

March 8, 2011 9:34pm UTC in response to Anand Arcot Narasimulu

Re: saw device comsol 4.0

Hi

I have some problems to follow you, as your initial model was missing a few points such as Elastic material and electric materials definitions for the Aluminium (which is NOT a PZT), then epsilonr of air is not defined (flagged red (!) ) etc

then eps.nJ[A/m^2] is the inward flowing current DENSITY so multiplying it with a boundary length of 1[um] does not give you Amps but Amps per 1[m] depth ([A/m] along Z out of paper)

and you have no voltage supply

--
Good luck
Ivar

Reply  |  Reply with Quote  |  Send private message  |  Report Abuse

March 9, 2011 4:15pm UTC in response to Anand Arcot Narasimulu

Re: saw device comsol 4.0

Hi

when you start with COMSOL, the biggest temptation is to run your complex model at once, this is mostly frustrating, as it is to go through all then examples, but from my experience its the best way:
- view a few videos on the main site
- repeat the exercices in the model library, print out the pdf files and follow the instruction, ask youself way this and that
- if possible, once you get a first catch on COMSOL, take one of the COMSOL courses these are excellent, allows you to get support on the fly and ask questions + hands on exercices, see the main site for dates and locations close to you. Check with your Comsol rep. If you are somewhat remote, at an university or so, find out how to group a series of users and find out with your institution how to go on, to invite COMSOL to your place, it's worth the try, in my opinion, but I'm not COMSOL, so I do not know their conditions and possibilities.

--
Good luck
Ivar

Reply  |  Reply with Quote  |  Send private message  |  Report Abuse