EMRFD Message Archive 2390

Message Date From Subject
2390 2008-11-11 20:56:20 chris_yipyap_com Figure 3.31 - single crystal filter
I'd like to use the circuit in fig 3.31 but at 10.000 Mhz instead
of 5.000 Mhz.

The 25.3 pF capacitor the text says "resonates the ferrite transformer"
and I can't figure it out. My calculation is that for 25.3 pF at
5.000 Mhz I need about 40 uH in there somewhere. But I don't see it.
So at 10.000 Mhz do I just make it ~6 pF instead of 25.3? Should I put
a little variable cap in there? If so, what indication would I be
looking for when I adjust it?

And I think I've figured out how to change the L-networks, so
no worries there, yet.

2391 2008-11-11 21:18:54 Wes Hayward Re: Figure 3.31 - single crystal filter
Hi Chris,

Golly, I had no idea that anyone would try to build it as it stands.

The inductor is bifilar. A FT-50-37 ferrite toroid has an
inductance constant of 68 nH/turn squared. So 12 turns will give 10
uH, so 24 turns will give you the 40 uH that you resonate with the 25
pF cap. For 10 MHz, I would drop to a smaller ferrite core and/or
use fewer bifilar turns. Drop down to something that will give you
around 2 or 3 uH for each side. You could probably do this with
powder iron at 10 MHz. The L-networks will have to be scaled to
something else. I went to 500 Ohms in this case. You may not want
to go that high at 10 MHz.

The transformer has one capacitor that resonates it. The other
capacitors (10 and the trimmer) serve to balance out the parallel C
of the crystal. The 191 pF caps are for the L networks that
transform up to 500 Ohms.

Find the crystal that you want to use and measure the motional L.
Then put that value into GPLA (or a simulator of your choice) but do
not include any parallel C with the crystal. You can then see what
value of terminati
2395 2008-11-12 07:07:18 chris_yipyap_com Re: Figure 3.31 - single crystal filter
Thanks for the quick reply!

I'm at the stage in my education where I take bits from here and
there, solder them together, then scratch my head when it doesn't work.

Right now I'm fooling with the free BF998 dual gate Mosfets and I'm
using a combinati
2397 2008-11-12 11:02:13 Wes Hayward Re: Figure 3.31 - single crystal filter
Hi Chris,

You may well have other issues to worry about too. You want a
narrow crystal filter, but you want it right at 10 MHz. If you grab
a typical crystal that is marked as 10 MHz, that means that it is
intended to generate 10 MHz in an oscillator. The series resonant
frequency is most likely 1 or 2 kHz lower than this. The single
crystal filter that you are building operates at the series
resonance. The parallel components are all cancelled.

What I would recommend is a simple ladder filter with just a couple
of crystals. But then insert series caps with each crystal to move
the frequency up to the desired 10.0000.

It must be interesting to be that close to WWV.

Good luck.

73, Wes

2400 2008-11-13 08:11:19 victorkoren Re: Figure 3.31 - single crystal filter
You mentioned receiving WWV and using it as a standard. It made me
remember such an application that I think you should see.
Check this link:
and this too:
I think that it might be just what you need, and with much lower
Victor - 4Z4ME

2402 2008-11-14 11:30:34 chris_yipyap_com Re: Figure 3.31 - single crystal filter
Thanks Victor!