Recommendations for Filter Synthesis/Ckt Analysis Software

[mcckevin]

All --
I teach a lab-based RF course and am thinking about adding a section on buidling and evaluating RF filters. We use Rigol DS815 spectrum analyzers with tracking generators as our network analyzers to test/tweak our filters after we build them.

I'm looking for recommendations for both the filter synthesis and circuit analysis apsects of the lab. I believe I can get student licenses of either Keysight GENESYS or Microwave Office from AWR, although I haven't yet looked into it. 

On one hand, I hate to give the students the idea that they need to buy a $10k+ software package to perform filter synthesis and circuit analysis when there are cheaper tools available. On the other hand, these are the tools they will likely use in a professional working environment. 

Less expensive options include LTSpice and the QUCS open-source circuit simulator. I haven't found much open source filter synthesis software beyond simple ham radio oriented material. 

So, any recommendations? Use student versions of the commercial packages? Cobble together (or find) useful pieces from the web? An integrated package would be ideal but I don't hold out too much hope.

Thanks.
Kevin

[T3sl4co1l]

I've heard NuHertz is good, but haven't tried it.

Simulators of course are fine for plotting and evaluating, but don't offer any analysis as such.

My, still fairly basic (and early), plotter at least gets out of the way, so you can make real-time adjustments:
https://www.seventransistorlabs.com/Calc/Filter1~.html
Mouse over the schematic to see buttons for modifying the network.  To "disable" a component you don't need, set a dummy value like 1e15 or 1e-15 as needed (open/short).

Tim

[KJDS]

You'll find that students can pick up and use Microwave office with about eight seconds of tuition, but anything from Keysight is a pain. I've never been a fan of Genesis, but it does have fans for it's filter synthesis tools. ADS is a far more powerful too, and I believe you can get student licenses for this too, but you'll end up spending more time teaching them how to use the tool that doing anything else though it is slowly getting easier to use and less bug ridden.

If it's degree level that you're teaching them to, then the filter synthesis should be done with a maths package.

[KE5FX]

I use AADE's filter design program, which I still think is the best of the free tools even though it hasn't been officially supported for a while.

Another popular choice is Elsie.  The free edition is heavily crippled, but the full package is only $95.  I don't know that it offers any advantages versus the AADE application, though.

[mcckevin]

I found two more of interest:

PCFilt at alkeng.com: I used this a long time ago when you had to pay for it. It's free now and very useful. He also has a couple of other programs that looked interesting but I haven't played with them.

In both class and professionally, I've used the free version of Arden Technology's SysCalc program to perform RF cascade calculations (cascaded noise figure, linearity, etc). I think the author passed on and his site (ardentech.com) has been taken over by someone else.

I really liked the older versions of EagleWare's (aka Circuit Busters) products, before they were bought up. They had good value, were interactive and intuitive. Unfortunately, my disks and keys have been lost to the ages.

[G0HZU]

Yes, the Eagleware suite was the best for filter design for many years. I have been using it for nearly 30 years at work and also here at home. The oldest working version I have here is the old DOS version of Circuit Busters' =FILTER= 2.1 and Super Star as below. However, I don't use this any more. I think I've got every major release of their SW from 1987 through to their last release of Genesys 2004.07 here at home.

I still use Genesys 2004.07 a lot. It has a few bugs and niggles but it is fast and easy to use and very powerful.

A free (but old) alternative is RFSIM99 as this has a basic filter design module within it. It can also simulate and plot the filter. It would be fine as a means to introduce a few classic filter topologies but it won't take long to reach it limits as a simulator.

[mcckevin]

Wow. I'm getting all misty eyed here.  I haven't seen those screens in ages! I remember running into a young and fresh-faced Randy Rhea at the microwave shows.

[w2aew]

I use AADE's filter design program, which I still think is the best of the free tools even though it hasn't been officially supported for a while.

Another popular choice is Elsie.  The free edition is heavily crippled, but the full package is only $95.  I don't know that it offers any advantages versus the AADE application, though.

These are the two that I thought of too.

[mcckevin]

Re: If it's degree level that you're teaching them to, then the filter synthesis should be done with a maths package.

This is a lab class. I really want them to rub their faces in realization problems (small value caps, large variations in component values, etc). The math is a good place to start but it often produces filters with impractical components. The ideal software would allow the students to apply various transforms (e.g. pi-tee, Norton, etc.) as a means of obtaining a realizable filter.

[mrf184]

We used RFSim99 at my university... and this was only 3 years ago. I still use it to this day plus LTSpice; very occasionally I go into my Windows VM and use Agilent ADS (if you look hard enough there is a crack; keep it in a VM and don't allow it internet access...) to verify a power amplifier design, but this is optional as you can do everything only using the impedances from the datasheet.

[mcckevin]

Hey G0HZU -
Any chance you could enable me to send you a PM? Thanks.
Kevin

[tfr000]

I use LTSpice for everything. Spice is quirky... I learned it back before Internet, so I'm used to it, but the learning curve is steep if you aren't familiar with it. LTSpice also has its own quirks.