Band Pass Filters.
 

Anyone have a design for band pass filters?

TIA

Cheers
Richard

Band Pass Filters.
 

Richard Neilsen wrote:
Anyone have a design for band pass filters?

Lots of people and lots of different design calculators on the web.

What kind of band pass filter did you have in mind audio, RF, LC, crystal,
cavity, strip line?

--
Jim Pennino

Remove .spam.sux to reply.

Band Pass Filters.
 

"K7ITM" wrote in message
...
If
you build a filter according to a design that either the AADE progarm
or Elsie gives you and it doesn't perform like you think it should,
don't be too quick to blame the program!

My experience is basically...

-- The behavior as-built will almost always be at lower frequencies than what
you designed for, due to the stray capacitances and inductances the simple
filters don't take into consideration. Similarly, bandpass filters will
usually end up narrower than you intended.
-- For bandpass filters, it seems just about as common that coupling between
elements, approaching self-resonant frequencies, and parasitics can help just
as often as they can hurt insofar as shape factor goes.
-- Taking a scrap piece of PCB material (or some other thin metal sheet-type
material) and placing it midway between pairs of inductors can give you some
qualitative idea as to how much coupling there is. (...and normally you want
to place consecutive inductors at right angles to one another... or perhaps
try the "magic" 54.7 degree angle like the old neutrodyne receivers used...)

If you need quite sophisticated filter designs, Nuhertz (aka, Jeff Kahler,
Inc. -- a man who can probably quote large portions of Zverev's book...) has
their Filter Solutions program, which is quite good... although only a very
basic version of it is free (apparently Jeff wants to eat every now and again,
and not just at Taco Bell).

---Joel "Yo quiero Taco Bell!"

Band Pass Filters.
 

On Mar 15, 7:48*pm, "Joel Koltner"
wrote:
"K7ITM" wrote in message

...

If
you build a filter according to a design that either the AADE progarm
or Elsie gives you and it doesn't perform like you think it should,
don't be too quick to blame the program!

My experience is basically...

-- The behavior as-built will almost always be at lower frequencies than what
you designed for, due to the stray capacitances and inductances the simple
filters don't take into consideration. *Similarly, bandpass filters will
usually end up narrower than you intended.

I'm glad Joel said "almost always." Be aware that shielding a
solenoid coil will drop its inductance (and its Qu), and thus raise
the resonance. So if you've designed a filter for a frequency well
below the self-resonances of the coils, and measure the coil
inductances without shielding, you may well find that the resonances
are higher than you expect when you put the resonators into shields.

Further to what Joel said: at least in Spice, you can add whatever
parasitic elements you want, and that can be valuable to understanding
why your real filter isn't behaving like the simple model said it
should. Recent versions of Elsie include the ability to write an
LTSpice output that opens into a nice schematic in the LTSpice
program, and lets you then play with modifying coil Q and adding
parasitic elements (including inductor-to-inductor coupling) to your
heart's content.

I agree that the NuHertz filter design software is very good, if
you're willing to pay for the full version. You can also get very
good filter design help in Agilent's ADS software, but they you'd
really have to live on rice and peanut butter for a looonnngg
time. ;-)

Cheers,
Tom