Earlier this month Paul VK3HN had a very interesting blog post about adding Automatic Gain Control to Termination Insensitive Amplifiers (TIAs).
Termination Insensitivity is especially important in bidirectional rigs. The shape of the crystal filter bandpass response is very dependent on the impedances presented at both ends of the filter. In bidirectional rigs you are changing the signal path direction through the filter when you go from transmit to receive. If the amplifiers at either end of the filter have impedances that vary depending on what is on the input or output of either stage, you will have great difficulty keeping the bandpass identical as you move from transmit to receive. Termination Insensitive Amplifiers let you do just that -- they stay at one fixed input or output impedance (usually 50 ohms) independent of what is attached to the other end of the amplifier circuit. This greatly simplifies impedance matching at the ends of the crystal filter.
When I started building BITX rigs, I asked Farhan about the impedance matching problem. He advised me to use TIAs on both ends of the filter and pointed me to a great 2009 article by Wes Hayward and Bob Kopski. Using the information from that article, I built my DIGI-TIA transceiver, and I have used TIAs in almost all of the rigs I have built since that project.
In his August 2021 blog post, Paul wanted to add Automatic Gain Control to the TIAs. He came up with a way to do this, but we worried that his circuit would have an impact on the impedance of the amplifiers.
Yesterday, Wes Hayward W7ZOI posted on his web site a TIA circuit that lets us do it all: Termination Insensitivity with Automatic Gain Control:
I now find myself tempted to rebuild one of the TIA stages in my Mythbuster transceiver, adding the AGC circuitry from Wes's design.
Thanks to Paul VK3HN for the blog post on this subject. And thanks to Wes Hayward for the TIA AGC design. It is a real privilege to have direct input from Wes on questions like this.
Thanks for posting on this Bill, and for your thoughtful analysis of the original idea.
ReplyDeleteI want to point out that the original idea was proposed, built and tested on a friend's uBitx by George VK4AMG. I was only the messenger!
George's post is in VK Home Brew on Facebook, March 3rd 2021. I won't attempt to post a link here, if you are on Fb, you should be able to find it, if interested. While you are there, drop a comment and like on some of the excellent VK homebrew projects.
It's an honour to have engaged Wes W7ZOI on this. I think this will be a most worthwhile extension to the ever-popular and easily reproduced TIA in our homebrew projects.
73 Paul VK3HN.
More cool beans from Beaverton. It just keeps coming . . . like the Energizer Bunny. I like how Wes ended this new paper: "There must be many other ways to obtain TIA characteristics. . . . FET circuits offer promise, owing to their high input impedance. There is ample
ReplyDeleteopportunity for the experimenter interested in circuit design at the component level." --Todd K7TFC