Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Electric Current vs. Electrostatics in Combination with Particle Trajectory or Chemical Diluted Species Modules
Posted 20 lug 2012, 15:51 GMT-4 1 Reply
Please login with a confirmed email address before reporting spam
Hello,
Please see attached paper for background on model I am trying to set-up. In the paper, they used chemical dilute species module and I am trying to use the particle trajectory module to figure out the position of the ions.
My questions involve the set-up of the electric field for this model. I have a 15mm long channel. The top boundary has an AC electric field with the shape of a bisinusoidal waveform with max amplitude of 1000V plus a DC offset of -20V. The frequency of the AC field is 2MHz, which means in order to solve for the right waveform, I need time steps of at least 5e-8 seconds or smaller. The bottom boundary is ground. My medium is air.
1) Since I have both an AC and DC potentials, do I have to use both electrostatics and electric current modules. If so, have I set them up correctly? If I don't have to use both modules, which one of the two would be the better one to use?
2) For a 15mm channel, with the flow rates I am using (Vin = 10m/s), it takes approximately 1 to 2 milliseconds for ions to reach the end of the channel. Using a time step of 5e-8, which is necessary to properly solve the AC waveform, the model has to solve 20000 points. This leads to really long computational times (~5 hours). Is there another way to model the bisinusoidal waveform not using a time dependent solver? or is there something I can do in linking the studies to reduce the computational time?
3) I also need to run a parametric study with the DC voltage applied to the top plate. My range is from -40V to 20V with a step of 5V. Is there a way of running the parametric and then using the parametric results as inputs for the particle trajectory? When I link the studies through the variables not solved for option, I can either tell the particle trajectory solver to use all the specified times or all the parametric values, but not both.
4) When I run the current model with a DC voltage of 0 and a time step of 5e-8 seconds, I get a convergence error at a time around 400e-5s. Could I get some help in resolving this? I could decrease the time step further, but that would just increase my computation time which is already very long.
5) When I solve the AC field for a time of 1e-3 seconds, and then use the field solution as an input for my particle trajectory, the particle trajectory solver seems to get stuck while compiling the equations. The solver keeps writing to my hard drive, until it has completely filled all the memory by writing temporary files. This problem is more prevalent if I use the chemical diluted species module and use the adaptive mesh solver. Is there something that can be done to avoid this problem?
6) My last questions refers to solving this model using the chemical dilute species module instead of the particle trajectory. As shown in the attached paper, the mobility coefficient for ions is modeled as a non-linear equation involving the electric field and some experimental parameters. If I use both the Electric currents module and the Electrostatics module, how do I make sure that both electric fields get incorporated into my mobility coefficient equation??
I really appreciate any help that you can give me.
Thanks,
Francy L. Sinatra
Please see attached paper for background on model I am trying to set-up. In the paper, they used chemical dilute species module and I am trying to use the particle trajectory module to figure out the position of the ions.
My questions involve the set-up of the electric field for this model. I have a 15mm long channel. The top boundary has an AC electric field with the shape of a bisinusoidal waveform with max amplitude of 1000V plus a DC offset of -20V. The frequency of the AC field is 2MHz, which means in order to solve for the right waveform, I need time steps of at least 5e-8 seconds or smaller. The bottom boundary is ground. My medium is air.
1) Since I have both an AC and DC potentials, do I have to use both electrostatics and electric current modules. If so, have I set them up correctly? If I don't have to use both modules, which one of the two would be the better one to use?
2) For a 15mm channel, with the flow rates I am using (Vin = 10m/s), it takes approximately 1 to 2 milliseconds for ions to reach the end of the channel. Using a time step of 5e-8, which is necessary to properly solve the AC waveform, the model has to solve 20000 points. This leads to really long computational times (~5 hours). Is there another way to model the bisinusoidal waveform not using a time dependent solver? or is there something I can do in linking the studies to reduce the computational time?
3) I also need to run a parametric study with the DC voltage applied to the top plate. My range is from -40V to 20V with a step of 5V. Is there a way of running the parametric and then using the parametric results as inputs for the particle trajectory? When I link the studies through the variables not solved for option, I can either tell the particle trajectory solver to use all the specified times or all the parametric values, but not both.
4) When I run the current model with a DC voltage of 0 and a time step of 5e-8 seconds, I get a convergence error at a time around 400e-5s. Could I get some help in resolving this? I could decrease the time step further, but that would just increase my computation time which is already very long.
5) When I solve the AC field for a time of 1e-3 seconds, and then use the field solution as an input for my particle trajectory, the particle trajectory solver seems to get stuck while compiling the equations. The solver keeps writing to my hard drive, until it has completely filled all the memory by writing temporary files. This problem is more prevalent if I use the chemical diluted species module and use the adaptive mesh solver. Is there something that can be done to avoid this problem?
6) My last questions refers to solving this model using the chemical dilute species module instead of the particle trajectory. As shown in the attached paper, the mobility coefficient for ions is modeled as a non-linear equation involving the electric field and some experimental parameters. If I use both the Electric currents module and the Electrostatics module, how do I make sure that both electric fields get incorporated into my mobility coefficient equation??
I really appreciate any help that you can give me.
Thanks,
Francy L. Sinatra
1 Reply Last Post 21 lug 2012, 18:49 GMT-4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
I can't give answer to all your questions, however if I am not mistaken:
1. If you have both AC and DC, you do not need to use electrostatics. Electric currents is sufficient for it. You can give an offset with initial values and you can also create an electric potential (or a terminal) to the part you're giving AC.
1. If you have both AC and DC, you do not need to use electrostatics. Electric currents is sufficient for it. You can give an offset with initial values and you can also create an electric potential (or a terminal) to the part you're giving AC.