KitsAndParts.Com Forum

The filter I'm using (roughly 800Hz) is shown in Spectrogram. The input to Spectrogram is the audio output of my transceiver fed into the mic input of my laptop. The transceiver is fed with a noise generator at the antenna input. Both the noise generator and an older (some say easier to use) copy of Spectrogram can be found at Tom's, N0SS's, website. Spectrogram as well as instructions (the are for a K2 but are applicable to any radio) can be found here http://www.n0ss.net/index_k2.html. The noise generator that I used can be found here http://www.n0ss.net/noise_generator_ac7ac-style.pdf.
I used 120pf with the 5Mhz crystals to make this filter.




*For a quick and dirty setup you could use band noise instead of a noise generator. Set a marker in Spectrogram for your sidetone frequency and adjust C5 on the Gilbert Cell Audio Detector board so that the sidetone is in the center of the display.
The markers here are 200Hz and 1000Hz. If the program allowed for more than two markers, I would have put another one at 600Hz

Last edited by ak2b (2008-09-25 15:13:43)

How do I match and grade crystals? Glad you asked

I built a test fixture where I can insert a crystal into a DIP machine socket.
Both  crystal leads are connect to a 330pF cap to ground and
both xtal leads goe to an I/O pin thru a 100 pF coupling cap.
Both I/O pins connect a 220 ohm load resistor to ground

I connect a scope to one end of the fixture and a sig-gen to the other.
Then I manually sweep the sig-gen freq looking for a peak on the scope.

Thus I find the loaded center freq and the series amplitude.
Those xtals with low feed thru get trashed.

-Diz

I started a project based on the schematic http://www.k8iqy.com/testequipment/pvxo/pvxopage.htm a few months ago. I have all of the surfacemount parts and started a board layout. I got sidetracked though. I'll finish it one of these days. I just don't have the time right now. I need to retire so I'll have more time. 23 months to go.
Leonard

The plot for this filter is much better on the Inrad web site. I didn't post this plot to compare differences (well, ok, price). I mainly wanted to see what a good store bought filter looked like when plotted on my VNA. I wasn't so sure that the cavity in the Universal Crystal Filter was caused by something I might have done. That's still possible but I don't think the problem is with the VNA.

Tom, ak2b

Interesting reading. Which brings up my question. I plan to follow DIZ's suggestion on the page for the crystal filters and use 470 caps. This will give me a 500 hz filter. After reading the above I'm thinking I may be starting out too narrow. Any suggestions?

Rick AA4W

Some questions about the crystal filter:

1. I would like to design filters for 2.1 and 6 kHz. Is this possible with this kit? My IF will be 4.9512 MHz. I need the motional parameters of the filter crystals to simulate the filter. Does anyone have them?

2. Since the MC1350 IF amp is designed to be fed from a 200 Ohm filter, I assume these crystal filter kits are 200 Ohms in/out, correct?

3. If the crystal filters are 200 Ohms and I want to feed the filter with the diode double balanced mixer kit, it seems the only way to mitigate the mismatch is to wire-in the attenuator on the output of the mixer, correct? If so, 3dB or 6dB or another value?

Thanks and 73's, David

drone wrote:

1. I would like to design filters for 2.1 and 6 kHz. Is this possible with this kit? My IF will be 4.9512 MHz. I need the motional parameters of the filter crystals to simulate the filter. Does anyone have them?

2. Since the MC1350 IF amp is designed to be fed from a 200 Ohm filter, I assume these crystal filter kits are 200 Ohms in/out, correct?

3. If the crystal filters are 200 Ohms and I want to feed the filter with the diode double balanced mixer kit, it seems the only way to mitigate the mismatch is to wire-in the attenuator on the output of the mixer, correct? If so, 3dB or 6dB or another valu1. I would like to design filters for 2.1 and 6 kHz. Is this possible with this kit? My IF will be 4.9512 MHz. I need the motional parameters of the filter crystals to simulate the filter. Does anyone have them?

2. Since the MC1350 IF amp is designed to be fed from a 200 Ohm filter, I assume these crystal filter kits are 200 Ohms in/out, correct?

3. If the crystal filters are 200 Ohms and I want to feed the filter with the diode double balanced mixer kit, it seems the only way to mitigate the mismatch is to wire-in the attenuator on the output of the mixer, correct? If so, 3dB or 6dB or another value?

Hi David,

I don't have the motional parameters for the 4 MHz crystals. I did use the AADE filter design program to design an SSB filter using the 5.06 MHz crystals. You can see the process about 3/4 the way down on page 6 of the 'Erector RX' topic. Diz's crystal are all pretty well matched (frequency wise) and by reducing the caps you can change the bandwidth. You can also use Spectrogram to do a quick and dirty scan of your crystal filter (see above  (this page)). It won't tell you much about deep skirts as it will about passband ripple - but it will get you by in the absence of a VNA. The SSB filter I built had a 1.5Khz input and ouput impedance according to the design program. So, I ended up using matching circuits for 50 ohms. I didn't use  Diz's IF amp. Any way, look over page 6 and you will see my trials and tribulations with filter building. It was well worth the effort - I learned a lot.
I would use the Wideband RF Amplifier after the DRM to provide a nice termination for the IF ouput port. You wire the amp for a 50 ohm or 200 ohm output impedance which would keep a following narrow CW crystal filter happy.
You can always go to the Elecraft site and download the schematic for the K2 and see how they deal with many of the issues that confront you when using the RF Toolkits. In fact, I'm in the process of shamelessly copying the agc of the K2 .

Tom, ak2b

My question is, can anyone answer his question about motional parameters ? Am I not correct in my thinking that they are different for each crystal and batch and he will need to make the measurements ? Or do I not understand the question (which is very possible)

kc0wox wrote:

Ok if it's show and tell time, I'll have to show mine I guess.

This is the test setup. The board is a prototype surfaxe mount bitx. The input is going into the if input anp, through an external 10 meg xtal filter and then through the  output if amp. This sets up aff of the filter impedances.

Nice looking board, is that your own design, or a kit?

wa2mze wrote:

kc0wox wrote:

Ok if it's show and tell time, I'll have to show mine I guess.

This is the test setup. The board is a prototype surfaxe mount bitx. The input is going into the if input anp, through an external 10 meg xtal filter and then through the  output if amp. This sets up aff of the filter impedances.

Nice looking board, is that your own design, or a kit?

It's a prototype of a surface mount bitx version 3. The circuit design is pretty much established but this is the first smd version. It's made by Elia from England. My board is about the second one made and we are running tests on it. The nice thing about it has a 10mhz if and there are no band determining components on it. With the addition of bandpass filters, vfo, amplifier, and low pass filters, it will work on anything between 180 meters and 6 meters.

It should match up well with Diz's bandpass and low pass filters and his 10 watt amp. I'll be hooking the combination up in the next couple of weeks.
Leonard

Thanks for the replies all. 73's David

* Hi Leonard, kc0wox,

Thanks for the info and pics, looks nice. Where did you get those little socket pins? Usually I rip the pins out of round machine-pin sockets or headers for that purpose.


* Hi Tom, AK2B,

Can you provide the values of the of the caps you used to make the SSB filter using Diz's kit and crystals? If so, also provide some detail on the circuit you used to match the filter's 1.5kOhm impedance to 50 Ohms.

I'll look at your post on the Erector_Rx thread. I have a miniVNA based on the IW3HEV design. See it here:

www.miniradiosolutions.com/

If I were buying a VNA today though I would go with the DG8SAQ USB Vector Network Analyser (VNWA) seen here:

www.sdr-kits.net

Significantly wider frequency range at about twice the price of the miniVNA. They're offering pre-built VNWNA units now; nice. Both the miniVNA and VNWA have active groups on Yahoo.

You said, "I would use the Wideband RF Amplifier after the DRM to provide a nice termination for the IF ouput port. You wire the amp for a 50 ohm or 200 ohm output impedance which would keep a following narrow CW crystal filter happy."

Hmmm... There's no clearly stated option with the Diz's RF amp for either a 50 or 200 Ohm output, there is a pi-pad there though and the kit comes with parts for a 3dB pad. I'm thinking the wider crystals are likely going to be mismatched too much without doing more than just a pad unless I want to toss a bunch of gain out the window. Also I'm thinking about an RF amp after the input RF filter and before the diode ring mixer that can be switched in/out.

I'll look at the K2 and see what they're doing. I too need something a bit more sophisticated than what I've got for AGC.

* Hi wa2mze,

Yes, regarding measuring motional parameters. Not too long ago Jack Smith, K8ZOA put up a some really nice papers on this subject. It turns out there are several techniques. It's not too difficult to do. Find the info on his site:

www.cliftonlaboratories.com

Again, Best 73's

This dude has a discussion about crystal paramters here http://www.qrp.pops.net/crystal-tester-2010.asp     and as soneone else said the AADE filter software has an excellent help file including tutorails in the help file

Drone wrote:

Some questions about the crystal filter:

1. I would like to design filters for 2.1 and 6 kHz. Is this possible with this kit? My IF will be 4.9512 MHz. I need the motional parameters of the filter crystals to simulate the filter. Does anyone have them?

2. Since the MC1350 IF amp is designed to be fed from a 200 Ohm filter, I assume these crystal filter kits are 200 Ohms in/out, correct?

3. If the crystal filters are 200 Ohms and I want to feed the filter with the diode double balanced mixer kit, it seems the only way to mitigate the mismatch is to wire-in the attenuator on the output of the mixer, correct? If so, 3dB or 6dB or another value?

Thanks and 73's, David