4x crystal ladder for ham receiver

[Garyl]

Hey,
I am in process on preparing parts for 20/40/80m receiver, and I need to find crystals for 4x 10MHz filter (with 47pF capacitors).
My question is, how crucial is the exact value of those crystals?
And from how many crystals do I have to choose from?
I can buy:
- 50 crystals US $0.16 / piece
- 100 crystals US $0.12 / piece
- 200 crystals US $ 0.07 / piece
Wwhich is the best options?

Also, do a crystal ladder filter must be made of crystals in HC49U cases? This is the "bigger" case which seems to be less popular nowadays....

 

[betwixt]

Crystals are usually very, very accurate so there is no need to pick and choose from them. If you only buy 50 you can make 12 receivers and still have some left over!

The case size doesn't matter but HC49U is the 'standard' for wire ended crystals. For home construction I would stay away from SMD packages as they can be difficult to solder and are more prone to damage if heated too long.

Brian.

 

[Garyl]

I've read that people are buying a bulk of crystals and choosing the four ones that have the "closest" frequency, but I don't know how crucial is that.

 

[SLK001]

I need to find crystals for 4x 10MHz filter (with 47pF capacitors).
My question is, how crucial is the exact value of those crystals?

And from how many crystals do I have to choose from?
I can buy:
- 50 crystals US $0.16 / piece
- 100 crystals US $0.12 / piece
- 200 crystals US $ 0.07 / piece
Wwhich is the best options?

Also, do a crystal ladder filter must be made of crystals in HC49U cases? This is the "bigger" case which seems to be less popular nowadays....

There is a small degree of tuneability with xtals that are slightly off, but you will have to know the f0 of each xtal. The 49U case isn't really all that large for 10MHz. Since the center freq of the filter will be lower than the f0 of your xtals, if you purchase 200, you might be able to specifiy a slighly higher f0 for the lot.

I'm assuming that you have some way to measure the individual xtal parameters?

 

[Garyl]

I'm assuming that you have some way to measure the individual xtal parameters?

I don't know what is the best way, but I was planning to use this simple crystal tester schematic:

along with simple PIC16F628A frequency counter, or maybe with the oscillscope.

I was planning to write down the measured frequency of each of them and then select the four closest values

 

[vfone]

Here is a nice (and old) crystal tester made specifically for crystal ladder filters:

https://www.robkalmeijer.nl/technie...hniek/hambladen/qst/1990/01/page21/index.html

 

[betwixt]

Post #5 schematic is just an oscillator using the crystal but I doubt with those capacitor values it will show it's true resonant frequency and in any case you would need an accurate counter to measure it.

A better way is to build TWO identical oscillators like that but on one of then change C1 to 180pF with a 40pF trimmer capacitor across it. Then using an oscilloscope in X-Y mode, connect one oscillator to the X input and the other to the Y input. Pick any two crystals from the batch and put one in each oscillator. It should show a loop on the scope trace, adjust the trimmer until it is as near stationary as possible. It almost certainly won't make a perfect circle but that doesn't matter. Note the setting of the trimmer. Now swap the crystals and repeat the process. Again note the trimmer setting. Finally, set the trimmer to the mid point between the two positions you noted, the oscillator parasitic capacitance should then be as closely matched as possible. Don't adjust the trimmer again! Try to keep the temperatures and environments of the two oscillators constant so one can't drift relative to the other but if possible run them from different supplies to minimize the chances of them injection locking together. You MUST fit a capacitor across VCC and GND (100nF ceramic) on both oscillators.

Now keep one crystal in it's oscillator and swap the remaining crystals into the other oscillator. The best matched crystals will show a stationary loop, the further apart their frequencies the more it will appear to spin. Pick the crystals with slowest spin as being matched.

This method doesn't check the frequency is exactly right but it does select the crystals nearest in frequency to each other. In a ladder filter, especially with a VFO driving the mixer stage, the exact frequency isn't too important but if the crystals are resonant at different frequencies the filter skirt will skew or in extreme cases refuse to let any signal through.

Brian.

 

[SLK001]

You're a HAM, so go find the article A Practical Test Set for Comprehensive Crystal Testing by Chris Trask / N7ZWY. I've had a .pdf of this for a while - it's from March 6, 2008 and it is on the web. It contains a relatively simple circuit for measuring all the xtal motional parameters that you need to design a working xtal filter using regular xtals. Without these parameters, building a 4-pole xtal filter is nearly impossible.