The AM Forum

Perhaps you Ol' Timers can remember certain high-voltage transmitting mica capacitors made by Aerovox, CDE, et al.  They were about 0.375" to 0.5" thick, had screw terminals about 1.25" apart, and were tan or brown colored.

 I've attached a (not very good) photo of one.

My question is this:  Do these caps have a higher or lower ESR than a modern COG disk ceramic at a frequency of about 2 Mhz. or in those days, 2 Mcs. ?  

Thanks for your help.

 Good question. It could be concluded that a low ESR capacitor would be good to pass a high frequency current with low loss. Then there are peak currents and RMS currents. A low ESR capacitor made for bypassing may not have an RMS current rating, so be cautious about using a leaded COG ceramic (also known as NPO) to pass a high RMS current unless the manufacturer has ratings to use the capacitor for such use. Some of the old big silver mica's did indeed have an RMS current rating stamped on the case, and this was usually at a specific frequency. Unfortunately those old capacitors do not always age well, as evidenced by leakage current, and perhaps as a consequence, have increased ESR.

Modern day capacitors with high RMS current carrying capability usually have copper straps for leads instead of wires. The link below covers this. At the link are several interesting PDF documents to download:
 
http://atceramics.com/Capacitors/18/rf-power-capacitor-assemblies

Jim
Wd5JKO

That style mica will take more RF current than the disc ceramic your comparing it against.   As Jim said, I am not sure how the absolute ESR between the two compares.  Due to the various construction schemes, some of the ceramics have very low ESR but as previously stated that does not mean they can withstand RF currents.  An exception is the CRL 850 series capacitors which are ceramic and they are commonly used as plate blocking capacitors and for that application they do have to handle rather high RF currents  

Here is a very good paper dealing with the various style capacitors and their suggested usage.  

http://www.qsl.net/i0jx/capacit.html

Joe - GMS    

I grabbed three vintage mica caps that have the same case style, and made some measurements at 2.0 MHz, using a VNA2180 network analyzer. Results:

.005uF 600wv measured 0.0054uF with 0.005 Ohm resistance, Q around 3000.

.01uF 1200wv measured 0.0105uF with 0.005 Ohm resistance, Q around 1500.

100pF 2500wv measured 101pF with 0.004 Ohm resistance, Q around absurdly high.

Use 'em.

John, W1TAG

I never heard what his application is.  Thanks for passing along your measured data.  Do those caps you measured have a current rating stamped on them?  Some do...

Joe GMS 

Joe,

No, they don't have any current markings. The G series (the ones with end bells) micas almost always do, as do the F4's (that's the "standup" version of the bakelite case caps). I've used this size .02 uF caps in filters at 137 kHz with 3-4A though, and have gotten away with it. No significant heat in hours of key-down. And the current capability derates with frequency, so it's a tougher test.

I didn't sweep up high enough in frequency to find the self-resonant points, though. That could be an issue past 10 MHz. Certainly no problem at the OP's 2MHz target.

John, W1TAG

Thanks Jim! 

Joe