AC current / circuit simulation software

[bbrianc]

I'm looking for software that can simulate an ac circuit. After a quick google, I found several free online services (which I will try out). What I was wondering though is if anyone had some advice concerning my selection. A brief list of requirements: accurate simulation of different conductor behaviours, high frequency simulation (high GHz range but under the THz range), skin effect and proximity effect analysis, ohmic heating, magnetic flux, reflections, magnetic permeability, phase shifting, inductance, eddy current generation, etc.

My hopes are to simulate a circuit as close to real world as possible before spending money on actual components. I also realize this may take more than one piece of software or I may have to program some of my own Matlab code.

As always, thank you all of you take the time to answer!

 

[berkeman]

I'm looking for software that can simulate an ac circuit. After a quick google, I found several free online services (which I will try out). What I was wondering though is if anyone had some advice concerning my selection. A brief list of requirements: accurate simulation of different conductor behaviours, high frequency simulation (high GHz range but under the THz range), skin effect and proximity effect analysis, ohmic heating, magnetic flux, reflections, magnetic permeability, phase shifting, inductance, eddy current generation, etc.

My hopes are to simulate a circuit as close to real world as possible before spending money on actual components. I also realize this may take more than one piece of software or I may have to program some of my own Matlab code.

As always, thank you all of you take the time to answer!

You can check out the list of "Similar Discussions" at the bottom of the page to see some recent discussions of SPICE software packages.

But it sounds like you are wanting to go beyond that, into a region where something like COMSOL might be needed. What have you found so far that will let you simulate the RF behavior of multi-GHz circuits? How high do you want to go in frequency? Do you have any design experience in the GHz range?

 

[bbrianc]

You can check out the list of "Similar Discussions" at the bottom of the page to see some recent discussions of SPICE software packages.

But it sounds like you are wanting to go beyond that, into a region where something like COMSOL might be needed. What have you found so far that will let you simulate the RF behavior of multi-GHz circuits? How high do you want to go in frequency? Do you have any design experience in the GHz range?

I'll be sure to look into those thank you. I had a feeling there would be similar discussions, but wasn't sure if they would be aimed in the same direction.

I never thought about a comsol / ansys solution but that's a very interesting angle! I'll have to look into that.

The closest thing I have found so far are more or less java applets and/or web based programs that don't seem to venture into the higher frequency range. As for how high in frequency, from my into AP calculations I don't foresee myself venturing past 1 THz at the absolute max.

I have no design experience in The GHz range. I'm actually an aerospace engineer by degree and a mechanical one by trade. Perhaps with a minor in EE by trade lol. I have worked in/on some electrical systems, but to this point it has been mainly DC.

 

[berkeman]

I'll be sure to look into those thank you. I had a feeling there would be similar discussions, but wasn't sure if they would be aimed in the same direction.

I never thought about a comsol / ansys solution but that's a very interesting angle! I'll have to look into that.

The closest thing I have found so far are more or less java applets and/or web based programs that don't seem to venture into the higher frequency range. As for how high in frequency, from my into AP calculations I don't foresee myself venturing past 1 THz at the absolute max.

I have no design experience in The GHz range. I'm actually an aerospace engineer by degree and a mechanical one by trade. Perhaps with a minor in EE by trade lol. I have worked in/on some electrical systems, but to this point it has been mainly DC.

Since you don't have GHz design experience, consider hiring an RF Consultant to help you with your circuit design. It's not something you can pick up and do simple simulations of to design and build a circuit. I'm on a project right now working at 2.4GHz, and even with 2 senior RF Consultants, it's a big challenge to get it working and in a form that can be certified by the FCC for use.

 

[bbrianc]

My limitations there come in the form of a DIY budget lol.

Luckily, at this point I have no need for FCC certification. I can imagine the challenge you must face.

I guess I can add that the circuit per se is not overly complicated in terms of number of components. The main complexity comes from the higher frequencies.

I greatly appreciate your advice though! Moving from theory to application always seems to carry new challenges.

 

[berkeman]

My limitations there come in the form of a DIY budget lol.

Luckily, at this point I have no need for FCC certification. I can imagine the challenge you must face.

I guess I can add that the circuit per se is not overly complicated in terms of number of components. The main complexity comes from the higher frequencies.

I greatly appreciate your advice though! Moving from theory to application always seems to carry new challenges.

Fair enough. Keep in mind that you can't just wire up an RF transmitter and start experimenting with it, though. The FCC (in the US) and other regulatory agencies have rules about what bands you can transmit in, and at what power levels. Many of these bands are licensed, and interference with adjacent bands is a bad thing. Picture accidentally shutting down EMS communication radios with your interference as you experiment...

If you do your experimenting in a good quality shielded room or enclosure, then that's usually fine. But if your experiments result in transmissions into the neighborhood or industrial park, that can earn you a visit from the guys with the antennas on their vans.

 

[bbrianc]

Wow lol yeah I'll keep that in mind! I'm in a semi remote area but I'll be sure to keep the levels under 1.21 gigawatts [emoji3]

 

[Baluncore]

If you know what you are trying to do you can usually get a prototype operating quickly by joining standard modules together with standard interfaces. It is a mistake to try to simulate and optimise everything to come up with an optimum solution early in the product lifetime. That way you will never build the prototype, you will end up optimising the idiosyncrasies of the simulator, trapped inside your computer.

You can never reach perfection, (the best). Perfection is the enemy of progress. You will produce a better solution, (second best), a year after the first prototype which must therefore be only third best. So the best you need to do now is only third best, to prove the concept, and you need to get it done quickly.

Come to think of it, because perfection is unobtainable, and a later improved model will always be possible, everything you ever do will only need to be third best. Don't simulate it, build it.