I have just designed and built up a 10MHz crystal filter. I am trying to measure the input and output impedance to see if it matches with the theoretical value of 158ohms AADE shows so I can match it correctly to my 50ohm RF amplifier. For testing and learning purposes I am trying to match this filter to 50ohms each end so I can connect it to my test equipment and minimise ripple.

I found a very helpful page as shown below:

http://www.eham.net/ehamforum/smf/index.php?topic=97625.0

The fourth reply down by Chuck WD4HXG looks to be what I need to do but I am a little unsure of the correct layout for the L resistor network.? Chuck says:

"Use an input L resistor network made with a 56ohm shunt resistor and a series 5K potentiometer between the filter and the 56 ohm resistor. Same thing on the output. Another resistor L network with the series 5K pot connected to the filter and a shunt 56 ohm resistor to ground."

Should the layout be as shown below (the in and out both have 50ohm impedance):



Should the potentiometers be connected as shown with only the wiper connected to the filter and the other end open and are the positions of the 56ohm resistors correct?

Thanks in advance for any assistance with this,

Kerr

Hi John,

I spent quite a while adjusting the resistors on my initial filter and it always seemed a little wobbly on the top, I did get a reasonable result eventually but I think my crystals were probably not the best. For testing purposes I bought a whole load of cheap ones so it would not matter if I made any mistakes as they were only a few pence each. You were right in that the resistance should have been around 200 ohms, I got about 260 ohms at one point and the other end was a bit higher.

I always like to try these things initially using cheap components so I do not need to worry about messing them up, once I know I am working along the right lines and I am getting roughly the results I expect I get better parts.

After doing this for a while I got distracted by Larry's videos (see previous post) and spent a while learning the very basics of SimSmith and working out a good LC matching network - this appears to have given me a reasonable looking filter which should be good for my 40m receiver I am designing.

My design is based on the one Charlie Morris is currently working on, I have just modified it to 40m:



Thanks again for the help and suggestions,

Kerr

Hi Gordon,

I bought quite a few (20) of the 10MHz crystals from Tayda, currently they are 10c each (perfect for experimenting with):



I spent quite a while going through them using a couple of different methods to sort them:

The first was using Ashhar Farhan's method as shown below:



specifically page 66:



The other method was the one from the following site:



specifically the 'Crystal parameters calculator' page:



I used these for my first Bitx20 and it works really well, I have listened (I just built the receive sections) to people from many different countries on it. When I built this receiver I did not have a spectrum analyser so I used an oscilloscope and RF generator to set up my circuits. Now I have a spectrum analyser I can look at my circuit blocks in great detail which is really interesting - I really like seeing how they all work and I can also tweak them until I am happy they are tuned exactly how I need them.

My other reason for creating this receiver (other than learning how to design each section from scratch) is learning how to use my new spectrum analyser to look at and adjust the individual circuit blocks (RF amplifiers, filters, mixers etc) - this is very interesting and I am really enjoying learning how it works.

I hope this helps,

Kerr