[TowerTalk] ICE-419a Bandpass filter

[jimlux]

Michael Tope wrote:
> Dave Hachadorian wrote:
> 
>> ----- Original Message ----- 
>> From: "Mike Schelly" <hinrgdj1 at comcast.net>
>> To: "'Towertalk Reflector'" <towertalk at contesting.com>; 
>> <ck2jones at yahoo.com>
>> Sent: Tuesday, July 22, 2008 3:24 PM
>> Subject: Re: [TowerTalk] ICE-419a Bandpass filter
>>
>>
>>  
>>
>>> The Cornell Dubilier DPPM20D1K from Mouser for 20M and if 
>>> the "coupling"
>>> cap is bad replace with the CDV16FF241J03...240pf, this 
>>> CDV16 series is RF
>>> rated! Also at Mouser.
>>>
>>> Mike K4EAR
>>>    
>>>
>> ----------------------------
>> I don't know how stable polypropylene would be at 14 MHz in 
>> a tuned circuit.
>>
>> The ones I've used successfully in the past are the mica 
>> CD15, CD19, CDV16 and CDV19, all on page 838 of the Mouser 
>> catalog. Use Mouser.com search on those terms, and more 
>> values will show up than are shown in the catalog.
>>
>> Dave Hachadorian, K6LL
>> Big Bear Lake, CA
>>
> 
> Yeah, Dave, if you compare the data sheets for the polypropylene and the 
> mica, the mica sure looks like it would be much better for high current 
> RF applications. That's not to say the the poly cap won't be good enough 
> in this particular application. Sounds like Mike K4EAR has tested the 
> poly caps in that application and found them to be adequate.
> 

There's a lot of difference between various polypropylene caps. The 
Tesla Coil folks use them in a high current RF application.  For 
instance, the 940P isn't nearly as good as the 942P series. You're 
looking for low ESR and extended foil construction, among other things. 
  Some PP dielectric caps are truly horrid, though.  The ones you're 
interested in are used in high frequency switchers, for one application.

One nice aspect is that in the event of a overvoltage, they're self 
healing to a certain extent.  Zap them enough times, and the C gradually 
decreases.

google for "tesla MMC" for more information.  They typically look for 
0.15 uF at 2kV, which they string together in 10-15 pieces in series to 
make a 20-30kV cap, but the general information on models that work is good.

A huge amount of destructive testing has been done on the various models 
and makes.  (One of the experimenters works or worked for a company that 
makes high power RF Plasma etchers and did lots of testing at 13.56 MHz, 
for instance)