QuantumATK Forum

General => Future Releases => Topic started by: serhan on December 9, 2008, 16:15

Title: Voltage sweep command would be useful
Post by: serhan on December 9, 2008, 16:15

Hi all,

First of all, congratulations for the new company and the forum!

I would suggest that if there were a command that sweeps the bias voltage from an initial value to a final voltage with some steps, and if the current versus bias voltage was written down in the *.vnl file then to be plotted, would be useful in the following versions.

Regards,
Serhan

Title: Re: Voltage sweep command would be useful
Post by: Anders Blom on December 12, 2008, 22:25

I agree it would be useful. For now, perhaps you can be satisfied with the script published in this post (http://quantumwise.com/forum/index.php?topic=19.0)!

Title: Re: Voltage sweep command would be useful
Post by: zdhlover on June 21, 2009, 10:04

Now I compute the I-V curve ,I found a question:using the script that you offered in Voltage sweep (I-V curve)(http://quantumwise.com/forum/index.php?topic=19.0) , I can compute the I-V by varying the voltage from 0 to ang value ,but I couldn't change the voltage from any value to zero, could your future release notice the problem that we can change the voltage not only from zero to other values but also can vary from  nozero value to zero in computing the I-V curve?

by the way ,how can I solve the problem now?

Title: Re: Voltage sweep command would be useful
Post by: Anders Blom on June 22, 2009, 09:23

There isn't really anything magical about sweeping the voltage. The script showed in the post you refer to can by tweaked extremely easily to support a different kind of sweeping. All you need to do, basically, is modify the "voltages" list. For instance, to run from 0 V to -3 V in steps of 0.5 V, use

Code

voltages = [0.,-0.5,-1.,-1.5,-2.,-2.5,-3.]*Volt

or for more flexibility, use NumPy:

Code

import numpy
voltages = numpy.linspace(0.,-3.,7)*Volt

and so on. You can just as well sweep backwards,

Code

voltages = [3,2,1,0]*Volt

but note that it can be hard to converge the high-bias calculation without initializing it from a lower-bias self-consistent density.

The updated version of the ivcurve script in this post (http://quantumwise.com/forum/index.php?topic=246.msg1438#msg1438) should also be of relevance, esp. for the case of starting from a high bias, which perhaps is converged by climbing up from zero bias.