0
\$\begingroup\$

I have made a circuit in LTspice and now I want to subject various nodes to stray radiation. So I want to add energy to those nodes by means of voltage or current sources (and probably some impedance network around them).

enter image description here

My first try was this, which makes probably no sense:

enter image description here

I added a voltage source with the impedance of free space.

enter image description here

It actually looks not so bad. The shield charges more at low frequencies and shield current increases at high frequencies. But on the other hand, it doesn't make sense for the radiation to have a reference potential, as there are no charge carriers involved. Also the effect should drop at low frequencies, as the radiation can't create DC charge displacement.

Questions

  1. Is this in any way a meaningful approach to simulate radiation on an exposed electrode? Of course, I would need to put my knowledge of the node shape/size into the description of the voltage/current source. But I cant think of how, a simple series resistance is probably not enough.

  2. How could the unit of the field strength (V/m) be related to the unit of the voltage source (V) ? Probably it has to do with the physical size of the simulated node in real space somehow ?

\$\endgroup\$
1
  • \$\begingroup\$ You may want to look into FEMM-related software. SPICE is good at certain things, and radiated electromagnetic waves is not one of them (though, maybe with great care and patience, it could give sime results). \$\endgroup\$ Apr 23, 2021 at 15:02

1 Answer 1

1
\$\begingroup\$

In LTSpice and the like you can't calculate/estimate conducted emissions nor radiated emissions of your PCB/product.


For radiated emissions you need a 3D electromagnetic simulation software.

That kind of software calculates E(x,y,z,ω) and H(x,y,z,ω) inside and on the surface of a user defined box that encloses your PCB.

Afterwards, through the Equivalence theorem it calculates E(x,y,z,f) and H(x,y,z,f) at any point in space outside the box.

f is treated as a parameter where the user specifies it's interval.

For example: 30 MHz < f < 1 GHz


More info here:

https://en.wikipedia.org/wiki/Surface_equivalence_principle


To model immunity of your circuit to external incoming RF fields, you have to model the external field E and H excitation using distributed voltage and current sources.

See here:

https://www.researchgate.net/profile/Changyul-Cheon/publication/263403050/figure/fig1/AS:296053577404419@1447596081508/Modeling-of-external-field-excitation-of-a-TL-using-distributed-voltage-and-current.png

LTSpice is not able to do it.

You need a 3D electromagnetic simulation software.

You will need to import your PCB as a 3D model file.

You will need to know and import the IBIS model of all I/O pins you need to monitor.

You will need to introduce in the 3D scene a radiating antenna that act as a source of radiation.

\$\endgroup\$
4
  • \$\begingroup\$ Thanks for the answer, but I don't want to simulate radiated emission. My question is about immunity. I want to check the effect of radiation on my circuit. So I want to add energy to those nodes by means for voltage or current sources (and probably some impedance network around them) \$\endgroup\$
    – tobalt
    Apr 23, 2021 at 15:19
  • \$\begingroup\$ Ok, I'll enhance my answer tonight or tomorrow. \$\endgroup\$ Apr 23, 2021 at 16:40
  • 1
    \$\begingroup\$ Ok, I'm done. I enhanced my answer \$\endgroup\$ Apr 23, 2021 at 22:14
  • \$\begingroup\$ Thank you Enrico. It looks like it is feasible at a fundamental level using current and voltage source to contribute the energy, but feeding the right amount in, is not readily done in spice. \$\endgroup\$
    – tobalt
    Apr 24, 2021 at 10:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.