0
\$\begingroup\$

In this answer it is said that the maximum capacitor current handling is "largely a mater of losses", what I can understand that is a factor for maximum dissipation.

So, what is the thermal impedances related to capacitors, in general electrolytic?

That is, does the case have a low thermal impedance to the inside, so a bigger capacitor will generally refer to best current handling? Or is the terminals? Taking the more available capacitors, some can say that being it or not, large current capacitors have large terminals, but not so relative to the current itself, I think. Does snap-in have more current handling capacity, or they price is higher just because of the mount type?

Of course if you have the data-sheet of the component you can take the parameters, but is not so common to find capacitors with data-sheets, or from capacitors taken from damaged equipments.

Also, does good dissipation of the capacitors can help both in current handling and MTBF?

\$\endgroup\$
2
  • \$\begingroup\$ Capacitor thermals will work in same way as other packages. ESR will decide DC power dissipation, and based Capacitors construction capacitors should have Thermal constant and maximum operating temperature. So, based on thermal constant and Maximum operational temperature of capacitor the maximum current of Capacitor depends. But most of the manufacturers will not give capacitor thermal constant, Instead they will maximum ripple current can be handled..... \$\endgroup\$
    – user19579
    Jul 3, 2014 at 6:49
  • \$\begingroup\$ @user19579 thanks, as I said, most capacitors "do not have data-sheet", or you get one from damaged equipments and does not have part number to search or can't find the data-sheet, that's why I ask if I can do an "estimate" by size, type of construction, capacitance, etc. In EPCOS data-sheets if I remember correctly, dissipation factor is specified. \$\endgroup\$ Jul 3, 2014 at 15:49

2 Answers 2

1
\$\begingroup\$

Large physical size will help dissipate power, but a large capacitor may also have large losses due to high ESR. Power supply capacitors sometimes have inverted ribs to increase the surface area and help them dissipate power (though I have not seen that in quite a while, Philips parts used to have them). Low-end capacitors are often physically smaller than those from reputable manufacturers for the same capacitance- avoid those where possible.

If you're scavenging stuff, you can try to look up the part numbers, or if it's no-name Asian stuff you might try to guess from the circuit function, but if the requirements are demanding, you might be better off to just go buy some parts of known specifications for a dollar or two. Generally even no-name parts will have the part capacitance, the working voltage, and a rated temperature marked on them.

A general rule is that the lifetime of an electrolytic capacitor will be at least 2,000 hours at the rated temperature (that's about a work-year). Some are rated for more life at the rated temperature. The temperature will almost always be marked on the capacitor- often 85°C or 105°C, but sometimes more.

The life doubles roughly for every 10°C you can reduce the temperature- internal heating via ESR losses contributes, but also external heating from other components and from the environment. If you want the thing to last a long time, use a high quality part and run it as cool as possible. If you want it to die fast, use a cheap part and run it hot.

\$\endgroup\$
7
  • \$\begingroup\$ Well, even some capacitors from reputable manufacturer (lets say Epcos) don't have part number marked on then. As I have some capacitors from scavenging stuff (well not to consider if its the capacitor itself the fault :) ) I would like to try to use. The use is for DC-Link capacitors for KWs SMPS. So I have some SMPS to get the capacitors, but I really don't think they will handle the current ripple or not. I have choose to reduce the individual capacitance and parallel the caps, so they share the current, anyway having some idea of the capacitor current capacity is a better idea. \$\endgroup\$ Jul 3, 2014 at 16:13
  • \$\begingroup\$ Anyway I think the larger physical size with less capacitance can give some guide of the relative current rating. \$\endgroup\$ Jul 3, 2014 at 16:15
  • \$\begingroup\$ For high-current (even let's say 3A peak), in my country at least, its a question of US$30,00/unit as a base, not US$2,00. \$\endgroup\$ Jul 3, 2014 at 16:21
  • \$\begingroup\$ will be at least 2,000 hours at the rated temperature (that's about a work-year). An year has aprox. 9,000 hours \$\endgroup\$
    – m.Alin
    Jul 4, 2014 at 11:21
  • 1
    \$\begingroup\$ @m.Alin because of the exponential temperature dependency time can almost stand still when many devices are not being used. So if you use it 8 hours per day, 5 days per week, that's the life. Of course a year is pretty short, so you might want to keep the temperature average to less than 50C (rather than 85C) during operation and have it last 10 years. \$\endgroup\$ Jul 4, 2014 at 12:16
1
\$\begingroup\$

It's sometimes helpful to think of caps as containing "charge storing stuff" and "current carrying stuff". In the case of a foil cap, it's easy to see the distinction: the surfaces area of the foil holds charge, while the bulk of the foil carries current. Making the foil thicker will cause it to have less surface area per unit volume, and thus have less capacitance, but the thicker foil will more easily carry current between the leads and parts of the surface which are distant from them. For electrolytic and other caps, it's harder to recognize the different parts, but the principle is the same. A cap with extremely thin electrodes could store more charge than one with thinner electrodes, but would have more resistance between the leads and the places that store charge.

The maximum current capacity of a cap is then limited by two factors: (1) the more resistance, the higher the voltage drop for any given amount of current; this will limit the amount of current that can be moved in and out of the cap without exceeding the maximum voltage; (2) pushing current through resistance will produce heat, and some caps are better able to handle the heat produced than others.

If a cap will never have to accept or supply very much current, using thinner electrodes will reduce the cost and space required to hold a given amount of charge. If it needs to supply more current, however, thicker materials will be needed, despite the higher production cost and increased size.

\$\endgroup\$
1
  • \$\begingroup\$ From yours first affirmations, yes that's more or less what I has thinking about the construction. So I think it's a bit ok to assume that large capacitors with small capacitance will have bigger current handling capacity as this is probably by thicker foil in electrolytic? \$\endgroup\$ Jul 3, 2014 at 15:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.