Measuring Crystal Parameters

[KI4YAN]

So I've been working on a crystal ladder filter-I got a nice box of crystals in a trade recently and have been trying to work out what to use them for.

They're 11.0592Mhz marked, and I have 50 of them. I'm pretty sure they're designed for microprocessors, as parallel resonant with 18pf load capacitors.

Anyway, I decided to use the G3UUR method of measurement, as it suited my test equipment better. I built a simple colpitts oscillator with some BC547 transistors, one as the oscillator and another as an emitter-follower buffer. Works fine, by far the easiest oscillator I've ever had to get going.



Only change is that I used 560pf capacitors in the divider, as that's what I had on hand. I also used a 47pf silver-mica for the switched in "detune" capacitor marked "Cs" in the above diagram.

Now here's the issue-I tested all 50 crystals and had three that would not run with the 47pf switched in. But all of them ran pretty close to 11.056XXXMhz. Again-not a big issue. Switch the added C in and they all run 11.058XXXMhz. Cool.

BUT, now I have to calculate Motional Inductance and Motional Capacitance...and for that I'm using the "Dishal" program by DJ6EV. Inputting my numbers outputs garbage. It seems to me that the swing from no series cap to series cap is way too big for the program to handle?

Examples of my data:

Crystal 1, Cs shorted with switch: 11.056516 Mhz
Crystal 1, Cs in series with xtal: 11.058436 Mhz
Motional inductance: 1381890.88 mH <---this is super wrong
Motional capacitance: 18.205 fF <---this is about double a reasonable value???

What am I doing wrong? Or did I just get a box of rubber crystals?

[KI4YAN]

Sorry guys, seems like I was just putting in too many decimal places-I measured to 1Hz, but the program doesn't like that many digits.

[G3UUR]

DJ6EV's program requires the frequency in kHz and I suspect you entered it in MHz, hence the very high value of motional inductance because it thought the frequency was 11.056kHz. There's no problem with specifying the frequency down to a Hz as long as you put the decimal point in the right place.

That value of Cm (18.2fF) seems fine to me - crystals on 11Mhz are pretty thin and with the normal area of electrode it could easily be that value.

The crystals that wouldn't oscillate with a high capacitive load (Cs shorted) would have been quite inactive anyway, so wouldn't have been suitable for filter use - too low a Q. You could use them for the BFO/CIO.

It seems like you've got all the makings of a good 6kHz AM filter for 11.060Mhz, so good luck with that!

73 Dave.

[Tom WA3KLR]

Jacob,

Here is some crystal data from my files for comparison:
Freq                   motional L  motional C   series R           Q                 Source
10.000 MHz fund. 12.7 mH       20 fF       7.96 Ohms     100,000      old McCoy CAT.?   HC-6 holder, TS-330 test set
10.000 MHz fund.  16 mH         15 fF       5                200,000        old Piezo Technology Inc. data sheet, AT cut HC-18 holder
16.000 MHz fund.  6.866 mH      14.4 fF   12.5          55,073 calc.    proprietary
 
Your motional capacitance seems correct.  The motional inductance seems correct if you move the decimal point 5 places to the left.  With your significant figures I get 10.034 MHZ series resonant frequency.  Perhaps if you try to measure the series resonant frequency yourself and use their motional capacitance and then re-calculate the motional inductance?

The calculated series resonance and parallel resonance oscillator frequency of your data is about 22 kHz apart at 11 MHZ; seems reasonable.

Good luck with your filters OM.

[KI4YAN]

Out of 50 crystals, once I lined up my error I was able to pick 5 sets of 6 that were within 50hz of each other, and a set of 8 within 30hz. Good batch of crystals!

Running them through the program, simulated a 6-crystal/110pf capacitor QER filter with a -3db bandwidth of 3khz (I know it'll be slightly narrower than that, should end up actually about 2.8Khz wide) and a termination impedance of 94 ohms. I was able to simulate the group of 8 crystals with 50pf capacitors to produce a 6.1khz wide filter with a termination impedance of 220 ohms. Great, there's my SSB and DSB filters, now to put them to use!

Now, 94 ohms, that's not bad, but the MC1496 mixer I'm using to generate DSB with claims the output impedance to be 40K and 5pF, at 10mhz. (I should be able to shunt the carrier null adjustment to generate DSB+carrier for AM, or regular DSB, or SSB, this way!)

How the heck am I going to match 40K to 100 ohms? that's a HUGE jump!

This is the circuit from the application note, that I'm using.

[KQ6F]

Add an emitter follower to the balanced modulator output.  Use a NPN transistor - connect base to pin 6 (eliminate the 0.1uf cap), collector to +12v, emitter to ground through 1k resistor.  Emitter to filter through 0.1uf cap.