tried using the NanoVNA to measure the crystal series resonance

 Got out the NanoVNA to try measuring the crystals.  It's almost scary to turn on the NanoVNA due to all the email traffic on groupsio - people have trouble with updating the firmware, connecting to other computer apps and myriad other items.  So I've left the firmware as it came, only "problem" is that there are "only" 100 sample points in a sweep.  So I zoomed in to  100 Hz  from 4,915,150 to 4,915,250 Hz.  This way each sampling point is 1 Hz so the resolution matches the ebay counter.

I viewed the W2AEW video on measuring crystals with the NanoVNA:  

 https://www.youtube.com/watch?v=G9zZRNzhsEE

I noticed the crystal fixture W2AEW used.  I tried something a little different (see it on the left)

using two male SMA connectors and a hunk of circuit board.  A Swiss army knife was used to cut the channels - kept cutting until the ohmmeter told me I had gotten rid of all shorts.  Pin sockets are used for the crystal connections.  I can see why people don't normally do something like this, even if the two connectors are perfectly aligned there is still a little strain on the unit during the attachment process.  And one thing with the NanoVNA is that I always seem to be horsing the something on and off the connectors.  

It would probably be better to do it the accepted way with a sacrificial male-to-female extensions and a flexible cable.

And I finally found a use for the extra stylus I had for my old Palm !  It works well with the NanoVNA touch screen  (it's the black plastic thingy above the NanoVNA).

The results:

  the NanoVNA readings correlated extremely well with the Butler grounded base oscillator.  The only crystals out of "position" with the NanoVNA were 9 & 4, they swapped places.  Now the readings are different but consistently roughly 80 Hz higher (+ 0.0016 %) readings from the NanoVNA must be due to calibration of the counter versus the NanoVNA.  Or maybe it's due to the Butler oscillator riding a slightly different point as the series resonance ?  When looking for series resonance wth the NanoVNA I just went to the middle of the dip on the reflective (S11) yellow trace.  The error isn't that great but the absolute accuracy isn't really important for sorting / matching the crystals.

Thinking about the error I tried a couple of things.  First, put the NanoVNA into the CW mode on 4.9152 MHz.  Second, tried measuring that signal with the ebay counter:  got 4.915218 MHz.  And also listened on the FT-450D: got 4.91520 (but using the Son of Zerobeat kit there is potentially some error).  Considering the data, since the nominal frequency of the crystals is 4.9152 MHz, the NanoVNA readings are 5 above and 5 below that point which seems ideal to me.  The 10 Butler oscillator readings are all below.  So I'd have to select the NanoVNA as the "winner".

I'm happy to finally get some confirmation on the Butler oscillator.  The G3UUR test oscillator circuit may have to wait, not sure if I have all the parts.

Best Regards,
Chuck, WB9KZY
http://wb9kzy.com/ham.htm