Message-ID: <4E467FE1.9080808@iup.uni-heidelberg.de>
Date: Sat, 13 Aug 2011 15:45:05 +0200
From: =?UTF-8?B?U3RlZmFuIFNjaMOkZmVy?= <Stefan.Schaefer@iup.uni-heidelberg.de>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; de; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
References: <4E4527EF.2000304@iup.uni-heidelberg.de> <52970A62D1224CAF9A966D1743DCBF00@PcMinto> <4E456AB7.6090403@iup.uni-heidelberg.de> <4FB899125601413A9077646E2728C784@PcMinto> <4E45A678.5010908@iup.uni-heidelberg.de> <4E45B0F9.1000007@iup.uni-heidelberg.de> <002101cc5947$0a6ce300$4001a8c0@lark> <4E45B929.70802@iup.uni-heidelberg.de> <F7DB147AC432438F8811F0E44DE2A13C@JimPC>
In-Reply-To: <F7DB147AC432438F8811F0E44DE2A13C@JimPC>
Subject: Re: LF: Re: Faulty FKP-1
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
X-AOL-SCOLL-URL_COUNT: 0  
x-aol-sid: 3039ac1d60cb4e4680dd4559
X-AOL-IP: 195.171.43.25
X-AOL-SPF: domain : blacksheep.org SPF : none


Hello Jim, tnx for comments.

Am 13.08.2011 02:28, schrieb James Moritz:
> Dear Stefan, LF Group,
>
> These metallised film capacitors are "self healing" - if a voltage 
> breakdown through the film occurs, the thin metallisation layers close 
> to the puncture are vapourised, clearing the fault, with a small 
> reduction in capacitance.
Yes, as written in the datasheet http://www.wima.de/EN/WIMA_FKP_1.pdf

> So they will survive a certain amount of over-voltage events. But this 
> can only happen a finite number of times...
Yes. In my case it seems the number of events was above the limit. And i 
expect the withstand voltage increases then since the arrangement 
becomes rather inhomogenious. And there must be a blow up due to the 
heat exposure, which is well visible in my case ;-)

>
> If you look at the manufacturers' rating data, they show graphs of 
> max. AC voltage vs. frequency. At low frequencies, the voltage is 
> constant, the limit being the value at which the dielectric breaks 
> down. Above some frequency which depends on the capacitor value, the 
> voltage decreases proportional to 1/f, equivalent to saying that the 
> capacitor current is the limitation at high frequency. If you work out 
> what this current is, it is often suprisingly low, and many amateur TX 
> circuits exceed the limits. But the criterion for the rated current is 
> usually an internal temperature rise of 10 degreesC above ambient, 
> which is quite conservative for a reasonably well-ventilated capacitor 
> in an amateur shack, so significantly higher currents are OK in practice.

So what do you suggest? Using the caps in series to bring down the 
voltage per cap or switching smaller values in parallel? I think it was 
an overvoltage problem so i will try the series arrangement.

Just realised that it is easy to calculate things: The LC tank circuit 
is directly connected to the PA output. So running 120 W RF at 50 Ohm 
means 1.55 A rms is passing that C. At 3.5 nF this means an applied 
voltage of  512 V rms. This was no problem at all during the last months 
(TXing in QRSS-60 and so). The caps are specified to 700 V rms at 1 kHz.
So when running 500 W RF, as i did during the tests of the last days, 
about 1 kV rms was applied. Surely this caused the breakdown. So if i 
design the caps for 2 kV rms, all will be fine :-) 6 A rms will be no 
problem for the caps i think, rather for the variometer ;-)

>
> Cheers, Jim Moritz
> 73 de M0BMU

73, Stefan/DK7FC