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.

COMSOLPIN photodiode example simulation (updated)

Please login with a confirmed email address before reporting spam

Hi, i'm trying to do a photodiode simulation based on the gaas PIN diode example from comsol. In this example, the electromagnetic wave is excited from port1 and port 2 at the semiconductor boundary, with uniform electric field. I made a slight change to the example. Instead of exciting a uniform wave at port 1, now i'm exciting it with a Gaussian field distribution, i.e, Ez=exp(-(x/1000[nm])^2). I've also made some change to the meshing(which works fine if the excitation is uniform). However, i'm getting this error:
Attempt to evaluate negative power of zero.
- Function: ^
Failed to evaluate temporary symbolic derivative variable.
- Variable: comp1.ewfd.normE@VDN${realdot@2}
- Defined as: (0.5*((realdot(comp1.ewfd.Ez,comp1.ewfd.Ez)+(realdot(comp1.ewfd.Ex,comp1.ewfd.Ex)+realdot(comp1.ewfd.Ey,comp1.ewfd.Ey)))^(-0.5)))
Failed to evaluate temporary symbolic derivative variable.
- Variable: comp1.semi.Enorm0_wo@VDN${realdot@2}
- Defined as: ((0.5*((realdot(comp1.ewfd.Ez,comp1.ewfd.Ez)+(realdot(comp1.ewfd.Ex,comp1.ewfd.Ex)+realdot(comp1.ewfd.Ey,comp1.ewfd.Ey)))^(-0.5))))
Failed to evaluate expression.
- Expression: d(((comp1.semi.Enorm0_wo*unit_m_cf)/unit_V_cf),{realdot@2})

anyone has any idea what might be the cause? Thanks a lot!

3 Replies Last Post 22 févr. 2017, 01:49 UTC−5

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 9 mars 2016, 09:16 UTC−5
is the negative power of zero due to meshing, convergence, or some other reason?
is the negative power of zero due to meshing, convergence, or some other reason?

Please login with a confirmed email address before reporting spam

Posted: 8 years ago 10 mars 2016, 12:41 UTC−5
any help would be appreciated, otherwise i probably will switch to TCAD altas to do this, which requires learning the new software...
any help would be appreciated, otherwise i probably will switch to TCAD altas to do this, which requires learning the new software...

Please login with a confirmed email address before reporting spam

Posted: 7 years ago 22 févr. 2017, 01:49 UTC−5

Hi, i'm trying to do a photodiode simulation based on the gaas PIN diode example from comsol. In this example, the electromagnetic wave is excited from port1 and port 2 at the semiconductor boundary, with uniform electric field. I made a slight change to the example. Instead of exciting a uniform wave at port 1, now i'm exciting it with a Gaussian field distribution, i.e, Ez=exp(-(x/1000[nm])^2). I've also made some change to the meshing(which works fine if the excitation is uniform). However, i'm getting this error:
Attempt to evaluate negative power of zero.
- Function: ^
Failed to evaluate temporary symbolic derivative variable.
- Variable: comp1.ewfd.normE@VDN${realdot@2}
- Defined as: (0.5*((realdot(comp1.ewfd.Ez,comp1.ewfd.Ez)+(realdot(comp1.ewfd.Ex,comp1.ewfd.Ex)+realdot(comp1.ewfd.Ey,comp1.ewfd.Ey)))^(-0.5)))
Failed to evaluate temporary symbolic derivative variable.
- Variable: comp1.semi.Enorm0_wo@VDN${realdot@2}
- Defined as: ((0.5*((realdot(comp1.ewfd.Ez,comp1.ewfd.Ez)+(realdot(comp1.ewfd.Ex,comp1.ewfd.Ex)+realdot(comp1.ewfd.Ey,comp1.ewfd.Ey)))^(-0.5))))
Failed to evaluate expression.
- Expression: d(((comp1.semi.Enorm0_wo*unit_m_cf)/unit_V_cf),{realdot@2})

anyone has any idea what might be the cause? Thanks a lot!


Please check initial values under Electromagnetic Waves, Frequency Domain study. I faced same problem due to zero (0) value under Electric field for x,y and z. This problem was solved using Ez=1
[QUOTE] Hi, i'm trying to do a photodiode simulation based on the gaas PIN diode example from comsol. In this example, the electromagnetic wave is excited from port1 and port 2 at the semiconductor boundary, with uniform electric field. I made a slight change to the example. Instead of exciting a uniform wave at port 1, now i'm exciting it with a Gaussian field distribution, i.e, Ez=exp(-(x/1000[nm])^2). I've also made some change to the meshing(which works fine if the excitation is uniform). However, i'm getting this error: Attempt to evaluate negative power of zero. - Function: ^ Failed to evaluate temporary symbolic derivative variable. - Variable: comp1.ewfd.normE@VDN${realdot@2} - Defined as: (0.5*((realdot(comp1.ewfd.Ez,comp1.ewfd.Ez)+(realdot(comp1.ewfd.Ex,comp1.ewfd.Ex)+realdot(comp1.ewfd.Ey,comp1.ewfd.Ey)))^(-0.5))) Failed to evaluate temporary symbolic derivative variable. - Variable: comp1.semi.Enorm0_wo@VDN${realdot@2} - Defined as: ((0.5*((realdot(comp1.ewfd.Ez,comp1.ewfd.Ez)+(realdot(comp1.ewfd.Ex,comp1.ewfd.Ex)+realdot(comp1.ewfd.Ey,comp1.ewfd.Ey)))^(-0.5)))) Failed to evaluate expression. - Expression: d(((comp1.semi.Enorm0_wo*unit_m_cf)/unit_V_cf),{realdot@2}) anyone has any idea what might be the cause? Thanks a lot! [/QUOTE] Please check initial values under Electromagnetic Waves, Frequency Domain study. I faced same problem due to zero (0) value under Electric field for x,y and z. This problem was solved using Ez=1

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.