[Hallicrafters] Capacitor self resonant frequency

[James Liles]

Hi Pete:
The reference to the shunt was from my original post --- had nothing to do with your notes Pete.  I think your method is great because you can view the effects over the entire spectrum of interest.  Problem is not everyone has a spectrum analyzer.  I had struggled with the use of the GDO to measure series resonance in caps and thought that I fond an alternative and inexpensive way to do it without the GDO.  That was my first post.  Used a signal generator to drive a variable tuned resonant circuit.  After resonating the tank inserting the test cap in series and re-resonated varying the frequency until inserting and removing the cap made no difference to tuning.  I found that the procedure was vulnerable to parallel resonance when the shunt was across the test cap.  The procedure asked you to resonate the circuit with the shunt in then resonate with the shunt out.  If the resonant frequency was different, change the frequency and repeat until the shunt in or out made no difference in tuning; purely resistive.  Problem was, I left the test cap in the circuit and applied the shunt to simulate it’s being out of the circuit.  I soon found that the shunt across the cap was parallel resonant at times very close to the series resonant frequency of the cap and appeared like a very low Q circuit even though I could get a peak from tuning.  Fix is to remove the cap when the shunt is applied.  

Kindest regards Jim K9AXN   

From: Peter Bertini 
Sent: Tuesday, January 04, 2011 2:22 PM
To: James Liles 
Cc: hallicrafters at mailman.qth.net 
Subject: Re: [Hallicrafters] Capacitor self resonant frequency

Hi Jim

I'm a bit confused, as normal :)  What "shunt" are you
referring to? 

The setup I used shows produces a pronounced
dip (inverted V) at the series reasonant frequency 
of the capacitor. It can be seen in the Photobucket 
photo referenced in my first post.

A few folks have raised some concern about the stub 
length introduced by the BNC tee introducing reactances 
that may have affected the accuracy of my readings. 
I wonder if adding a few more BNC adapters after the
Tee would show a significant shift in the apparent
resonant frequency?  Better yet, I should go back
Iand do this again I will try using a section 
of strip line so the cap leads can be directly soldered 
on the 50 ohm line without involving a Tee fitting.

I didn't have much luck coupling into my Millen
GDO. More experimentation is needed there.
I suspect my AEA analyzer could be used to
verify the readings, but the LCD display has failed 
and I need to replace it.

Pete


On Tue, Jan 4, 2011 at 4:01 PM, James Liles <james.liles at comcast.net> wrote:

  Good morning Pete, Richard and all!
  Great links Pete – didn’t know just how much I didn’t know about caps!  I tried your setup with the spectrum analyzer although I don’t have a tracking generator.  I used a 10db attenuator to a “T” to the spectrum analyzer.  At the “T”, I wire wrapped a lead on a .001 20% 1Kv  disc ceramic measuring 3/8” and slipped it over the pin and went straight to shield with the other lead; very short leads.  Set the spectrum analyzer to linear and max hold and swept in 10 meg segments.

  Results were a decline to 59Mc and then an incline.  Didn’t try switching out the “T” for the adapter in the procedure that you sent; the measurement will probably be more accurate with that configuration.

  Now, because others were able to extract indications with a GDO, I decided to revisit the effort.  I could not attain a definite dip without some form of a loop in the leads, and the leads and shape seem to be responsible for the majority of the inductance.

  I wound a new coil measuring 1/4” in diameter for the GDO to concentrate the field.   Then took the same .001 cap and clipped one lead leaving but a stub and wrapped the other lead directly around the stub leaving a 1/4” wire between them, with almost no space between the lead and cap.  Directed the dipper coil directly at the point where the lead and cap meet and was able to distinctly measure a dip at 81Mc.

  Now regarding the first post that I sent.  I thought it was a great idea for a low cost method to measure self resonance in capacitors but the GDO method is easier to use and appears accurate ---- to be determined!

  One unexpected result with the setup in my original post is this.  I said to connect the test cap to the circuit and leave it there and connect one side of a shunt, leaving the other free to connect and disconnect.  There’s a problem Houston!  With the shunt connected, the shunt and test cap become parallel resonant under some conditions and the whole of the circuit series resonant causing the circuit to appear to be very low Q.  The test cap must be removed when the shunt is applied.  Am I going to use this method in the future?  NO, the GDO!

  Kindest regards Jim K9AXN
  ______________________________________________________________
  Hallicrafters mailing list
  Home: http://mailman.qth.net/mailman/listinfo/hallicrafters
  Help: http://mailman.qth.net/mmfaq.htm
  Post: mailto:Hallicrafters at mailman.qth.net

  List Administrator: Duane Fischer, W8DBF
  ** For Assistance: dfischer at usol.com **


  This list hosted by: http://www.qsl.net
  Please help support this email list: http://www.qsl.net/donate.html