[Hallicrafters] Capacitor self resonance

[Richard Knoppow]

----- Original Message ----- 
From: "Peter Bertini" <radioconnection at gmail.com>
To: "Richard Knoppow" <1oldlens1 at ix.netcom.com>
Cc: <hallicrafters at mailman.qth.net>
Sent: Sunday, January 02, 2011 8:34 PM
Subject: Re: [Hallicrafters] Capacitor self resonance


I thought I would offer up my findings for series resonance 
frequencies for
several caps I tested in the shop. I didn’t have much luck 
with the grid
dipper method, so I emulated the following HP setup as close 
as I could:



http://www.home.agilent.com/upload/cmc_upload/All/exp79.pdf?&cc=US&lc=eng



Equipment used was one of the shop’s HP-141T generators with 
the 8553B RF
deck and companion 8443A tracking generator. A photo of the 
bench setup and
improvised test jig is here:



http://i117.photobucket.com/albums/o46/radioconnection/DSCF1043.jpg

If the premise for this setup is flawed, I'd appreciate 
feedback.


I didn’t have a large selection of disk ceramics, but here 
is what I
measured.


    This is an interesing paper, thank you for posting the 
link. This test is at quite low impedance, 50 ohms, but 
should be valid for others.
    I have been researching self resonance and have been 
surprized by how difficult it is to find more than 
theoretical material defining the effect. Supposedly 
manufacturers publish self resonance data but I've been able 
to find it only for surface mount caps meant for microwave 
use. This have self resonance in the Ghz range.
    The inductance with which the capacitance resonates 
appears to be more than just the leads. Some types of 
capacitors must have fairly high inductance due to the 
construction of the capacitor.
    I've found data on series inductance for some General 
Radio standard caps. These appear to have self resonance of 
a few mhz. They are of the mica stack type so are not 
typical of caps normally used for coupling and by-pass 
purposes.
    I did find a link to an IEEE paper on measuring self 
resonance but they want a lot of money to download it unless 
you are a member. I let my membership lapse years ago.
    BTW, in the General Radio catalogues it mentions that 
the variation of capacitance with frequency is greater for 
mica dielectric than for polypropylene at DC and low 
frequencies so they are recommended for standards at audio 
frequencies. This might also be important for coupling 
purposes and for filters where exact value is critical. The 
variation is not large but its there. Most of the GR 
catalogues printed from about the late 1950 and later have 
the infomation in them. The reason given for the variation 
is interface polarization of the dielectric, which has a 
time constant. This is a subject that will get you right up 
to the eyebrows in vector and tensor calculus and partial 
differentials. Capacitors are a lot more complex than most 
people think.
     I am not at all sure of the results I got using the RX 
meter because I am not quite sure what I was measuring. 
Capacitors can have more than one resonance so I may have 
been fooled. The reactance range of the meter for both 
inductive and capacitive reactance is very limited so 
measurments of capacitors beyond its range are meaningful 
only near resonance.




--
Richard Knoppow
Los Angeles
WB6KBL
dickburk at ix.netcom.com