Dear all,
This is a topic like the audio "quality" of copper speaker wire
that usually leads to lots of opinions and folklore. I have not followed
most of the thread so this may be repetitive:
* At the time of manufacture electrolytic capacitors are fromed by slowly
raising the voltage to a value something like 10-15% above the nominal rated
value. Under these conditions, the capacitor forms a protective insulating
layer on the surface of the aluminum foil used for the plates. During the
forming process the capacitor draws much larger leakage current than normal.
The thin layer acts as dielectric/insulator of the capacitor. The foil has
been first treated to increase the effective surface area by etching or other
physical means. The ever incresing volumetric efficiency of modern capacitors
has been achieved by thinner foils and proprietary techniques for the foil
etching. Photoflash capacitors are somewhat different from other electrolyte
capacitors in that they are designed to take very large peak currents in the
order of 200-300Amps that flow through the flash tube. The high current
rating is achieved by making the connection to the plates all along the foil
edges continuous, rather than only with a few foil tabs which could be
vaporized by such high currents. Photoflash capacitors are also rated in a
very different way: Instead of being rated for a certain number of operating
hours at rated voltage and temperature, they are rated in terms of a certain
number of flashes. For small on camera flashes this may be something like
1500-5000 flashes. For large general purpose capacitors they are often also
rated for a brief (few seconds) "surge voltage" typically about 10% above
rated. Large capacitors used in studio flashes are rated more like
traditional capacitors in operating hours, since they are operated for long
periods with voltage applied. The volumetric efficiency of those devices is
much lower than on camera flashes. Interesting in auto and TTL flashes the
semiconductor switches (SCR's etc) used are also rated at only a few thousand
flashes at 200A! (see the Mitsubishi website for example data sheets).
When capacitors are operated at rated voltage and espcially at high
temperature, leakage current increases causing more self heating and
shortened life or even thermal runaway. The leakage current drops off rapidly
below rated voltage, so in conservative designs parts are derated and
operated below rated voltage to extend life. For photoflash units this is not
usually done because of extreme size constraints. Operating capacitors well
below rated voltage slowly deforms the capacitor this changes the passivating
layer thickness (voltage) and usually causes an increase in capacitance at
the same time. In electronic circuits Tantalum electrolyte capacitors are
very popular because of good high frequency performance. These devices are
particularly prone to shortened life at rated voltage so experienced
designers usually derate the voltage substantially and companies like
Panasonic even recommend this. For example I almost always choose devices
with 50-100 0gher voltage than I need. (eg 10-12V devices in 5V digital
circuits ) When left off for extended periods (years ) capacitors deform
leading to the classic explosion on switch on. Old vacum tube radio
collectors thus often reform the capacitors by connecting the radio to a
"Variac" transformer and slowly (minutes) increasing voltage applied. Small
photoflash units are inherently relatively slow to charge except with a Metz
Tubo unit maybe, so in some sense they are kinder to the capacitors. You
could also use used batteries to slow the rate of rise of voltage, if you
have not used the flash for a long time. Recharging the flash every month
seems very excessive despite the exhortation of some of the flash vendors.
Not flashing the flash for a few minutes after first operating the unit after
a long break, probably is a good idea, because self heating from the flash
currents will then not occur until the capacitor has reformed itself.
Regards,
Tim Hughes
>>Hi100@xxxxxxx<<
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|