Baby's First PCB

 Along the lines of making an SSB, I thought using bi-directional amplifiers would simplify things (as in the BitX).  I decided to make them termination-insensitive as in W7ZOI's design.  As I would end up making several of these, and they seem pretty useful building blocks, I decided to dip my toe into PCBs for the first time.

I decided to try JLCPCB, as they have very inexpensive prices *and* I could get them manufactured there, which would allow me to make them smaller and use small SMT parts that I wouldn't normally use (as by the time in the day I can work on my projects, my eyes do not like that...)

I got some really great advice on the EEVBlog forums.  For one, I found searching through the available parts *really* hard, and someone suggested this project that scrapes JLCPCB's parts list and presents a *much* nicer search interface.  One of the main things is it is much easier to find "basic" parts, which don't cost you more to add ("extended" parts cost $5/part since someone has to manually load them).

I used EasyEDA and followed one of Electronoobs YouTube tutorials.  After some back and forth on the design based on feedback from the forums, I ended up with this:

Ordered it, and a week later I had this

Very pleased.  Hooking it up and measuring it on the oscilloscope (into a 50 ohm load), it looks like I am getting 12-13dB voltage gain at 7Mhz from around 10mV up to 272mV without distortion.  I need to go back and look at my notes - seems lower than I was expecting.  Also want to measure with the NanoVNA (for bandwidth) and the TinySA (for harmonics).  But that will probably need to wait for another day.  Here are a bunch of pictures:

Baby's First Crystal Filter

 I've decided I want to make an SSB, along the lines of the BitX or something.  Along those lines, I decided I should make a crystal filter for use in it.  I'll be following the directions from W7ZOI.

First, I ordered a lot of 100 9MHz crystals off of Aliexpress:

I plan to match them using the NanoVNA.  The Windows software for that has instructions on making a test jig (basically 4:1 matching transformers and 3db pads on either side of the crystal) and apparently has a mode to let you match crystals in batch.  But I got impatient and wanted to see what it would look like to just make one with the first four crystals I pulled.  The result is here

And the sweep on the NanoVNA as follows

The best I can tell, it is centered at 8.9978MHz with a 1.8KHz bandwidth (to the -3dB point), with a 2.48dB insertion loss.  Which isn't far off what I want (was aiming for a 2KHz bandwidth).  So all in all, a pretty good result I think?  I want to try again with NanoVNA Saver when I get a chance (to get better numbers/pictures) and I really should actually calibrate it :)  But it was a quick test when I had a few minutes between things. 

Edit: Got NanoVNA going, and calibrated.  I was pretty spot on before.