Haven't electrolytics gone through some changes in recent years? 40 years ago they were not very good and tantalum caps were plentiful, available and not connected with bad politics. Since then, hasn't there been some progress in density, operating temperature and ESR of aluminum electrolytics? I would like to know more about that.
For the last 40 years, I never used aluminum electrolytic capacitors, for the reasons stated in this article.
I was always using tantalum capacitors. Yes they are more expensive! (By how much these days? I am now retired...) And you know the old adage about how much does the cost to replace a $1 part escalates as we progress from concept to the field of the product's life cycle...
Yes you have to pay attention to the voltage rating, the temperature rating and the ESR, but in order to get long life out of an electrolytic capacitor is to make sure that you are operating below the ripple current rating. It may be necessary to use lower value caps in parallel or to use an electrolytic with a higher ESR in parallel with a ceramic or other cap with a lower value and ESR to minimize the ripple current in the electrolytic. Inexpensive electrolytics tend to have a lower ripple current rating and physically small capacitors with a relatively large value also tend to have a lower ripple current rating.
True in general more than specific. SMPSs such as in computers, and these days motherboards (think POL switching converters) are also vulnerable. In all of these cases, though, I think is has less to do with voltage derating than ESR. Low ESR capacitors are more expensive, but they last longer because the produce less temperature rise with ripple. The extreme case is the DIY who replaces them with same capacitance and voltage from Radio Shack, without realizing that a low ESR device is required. That can lead to spectacular failures in a matter of hours.
Companies should make these electrolitics easily replaceable with detachable or sub assemblies that can be replaced easily when the cost is prohibitive to use other technologies yet the electrolitics are an order of magnitude worse in failure than the rest of the system is.
Same issue with all the flat screen tv's you see out on the curbs dead.
its the electrolitics.
Most of these do not get exposed to external sources of heat like this LED example shows.
But they die in the same overtemperture boiled off electrolyte way. They were not deraited for voltage and thus run hotter when driven near 100% of voltage raiting. The AC ripple voltage has a higher RMS power drop in the capacitor at higher and higher voltages. This is why the general rool of thumb is derate to 50 or 40% of rated voltage so that the MTBF of the capacitors has a chance to last as long as intended.
The only thing more unreliable than an electrolitic capacitor even when derated is a battery.
A Book For All Reasons Bernard Cole1 Comment Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...