Inrush limitation is to prolong electrolytic cap life?

[cupoftea]

Hi,
Page 5, point A, of the following says inrush limitation is needed to protect the mains rectifier diodes...but surely, it is rather, to protect the input electrolytic capacitors?

https://www.power.com/sites/default/files/documents/linkswitch-tn2_family_datasheet.pdf

 

[betwixt]

A bit of both and to stop the input fuse popping as well. Remember a capacitors ESR will limit the charging current to some extent but the diode surge rating may only be for a few mS.

Brian.

 

[crutschow]

Page 5, point A, of the following says inrush limitation is needed to protect the mains rectifier diodes...but surely, it is rather, to protect the input electrolytic capacitors?

Why do you think that?
Diodes can be destroyed by high surge currents, whereas electrolytics have little sensitivity to that.

 

[Easy peasy]

Indeed - many 400 & 450V electro's can handle well over 100A without issue ( 470uF ) - we have done a fair bit of testing of this for both charge and discharge - which are essentially the same thing as far as the effect on the cap is concerned.

The main thing an inrush limiter does is add to the R of the mains so that the over voltage ring up is limited at switch on, for very low R the theoretical max ring up is 2x Vin peak.

 

[cupoftea]

Thanks, i used to repair streetlights....never saw a blown diode bridge...saw loads of blown fuses.
ive been amazed at how resilient mains Si diodes are too massive overcurrent...used to send them for surge test...never did mains bridge blow.
Electrolytic caps are delicate, the liquid electrolyte will have cumulative chemical reaction on inrush current heating...sorry just seen we posted same time.

Good point about ring up...and of course, that is the electro caps that are going to be most damaged by that...as they are only 400v usually.

 

[Calm down]

Surge refers to the cold shock when starting the machine, which is easy to produce sparks on the plug, and at the same time, it produces instantaneous overcurrent on the fuse, bridge stack, and electrolytic capacitor, which is also the reason why many fuses use time delay insurance. Small bridge stacks are more sensitive to impact, and using diodes can be much more reliable or using larger bridge stacks. The impact of capacitance actually depends on the withstand voltage of the anode aluminum foil. The 400V capacitor anode aluminum foil may be 480V, 510V, or 530V, and the higher the voltage, the more expensive and reliable it is. There are also capacitor leads and internal aluminum foil that are pressed together, which can easily ignite or break when encountering impact if not done well.

 

[KlausST]

Hi,

Charging a capacitor needs 0.5 x C x V x V
And in a non inductive circuit an extra 0.5 x C x V x V is dissipated as heat. Independent of series R.

Example (ideal devices)
Charging a 100uF to 50V with an ESR of 50mOhms will result in the same "waste of energy" as charging it via 1000 Ohms series resistance. Same amount of heat dissipated.

Electrolytics lifetime mainly depends on temperature. So adding a series R will not reduce the total heat. And as long as capacitor and resistor are not thermally isolated .. the extra R won't reduce temperature. Indeed it rather will increase overall power dissipation, because DC current has no influence on C power dissipation, but will increase power dissipation in the R.

Klaus

 

[danadakk]

Not an expert here but Electrolytic lifetime highly dependent on its T, as previously stated.
The C gets rid of its heat thru radiation, conduction, etc.. This is classic thermal mass T decay
problem.

So is it correct to say that lower charging rates result in lower T rise in C ?
As C thermal mass decay allows multiple time constants to get rid of C
heat ? Versus dumping high charge in very short time constant into C and
decay not able to utilize time to get rid of heat energy proportionally faster
as the defining decay system is passive, fixed properties.

Obviously in the limit C temperature for ESR = infinity = ambient T. Versus dumping
a lot of energy into it in a very short period of time and waiting for decay to remove it.

Temperature Change and Heat Capacity | Physics

courses.lumenlearning.com

Regards, Dana.

 

[KlausST]

Hi,

So is it correct to say that lower charging rates result in lower T rise in C ?

It depends on what you are focussed on.
--> If you are focussed in the capacitor temperature only then you are correct.

But usually the capacitor belongs to a circuit and the circuit maybe is installed in a case. Now the temperature inside the case obviously does not only depend on the capacitor´s power dissipation, but also on all other parts around dissipating energy.
--> If you are focussed on the whole application, then the lower charging rate does not necessarily reduce temperature.

***
If we talk about a "one shot" charge of a capacitor at power on ... then about what temperature rise do we talk?
Temperature rise at low and high charging rate.. does it influece lifetime significantly? I doubt it.

***
If I´m not mistaken a temperature rise of about 9 degree will reduce lifetime to about 50%.

Klaus

 

[cupoftea]

Thanks, as you know, the inrush will be in some way limited by the caps ESR....and considering say, the 860021374027 capacitor by Wuerth, like other Electro caps, there is no realistic way of calculating what is the ESR that the inrush surge will "see".....you have to guess it will be something like 0.1R (the datasheet gives DF, but at 830MHz!)......that 0.1R will dissipate quite a bit from the inrush surge.........which is unwanted heat inside that electrolytic cap can.

https://www.we-online.com/components/products/datasheet/860021374027.pdf

--- Updated Apr 28, 2023 ---

As you can see in the attached LTspice and jpeg, the inrush into the 22uF cap, causes 170W of dissipation in the electro caps 0.1R ESR...for some 190us pulse.....this isnt likely to do that cap any good, especially if it happens several times a day...for many years.