Electrolytic capacitor voltage

[olly_k]

ok so we all know that it is normal to choose an electrolytic capacitor with maybe 15% voltage headroom or something similar, but is there a point at which a high voltage electrolytic capacitor will suffer if it is subjected to a far lower voltage than it's rating?

 

[Sceadwian]

I've never encountered any lower limit, the only difference between a lower voltage and higher voltage electrolytic cap is the thickness of the dielectric material, there may be some changes in electrolyte chemistry or general internal design with higher voltages but this shouldn't affect lower voltages even down to single volts.

 

[crutschow]

I know some old electrolytics had to be "reformed" by slowing applying a voltage to them if they had been unused for a long period of time, but I don't think modern units suffer from that problem.

 

[BrownOut]

ok so we all know that it is normal to choose an electrolytic capacitor with maybe 15% voltage headroom or something similar, but is there a point at which a high voltage electrolytic capacitor will suffer if it is subjected to a far lower voltage than it's rating?

You should try not to go lower than about 75% of the rated voltage. Some electrolyitcs can degrade if the working voltage is too low.

 

[tcmtech]

I don't know how a lower voltage limit could be possible, other than a reverse voltage connection, being in most circuits capacitors spend large parts of their service life in zero volt conditions.

You know, when the device is turned off.

 

[killivolt]

I don't know how a lower voltage limit could be possible, other than a reverse voltage connection, being in most circuits capacitors spend large parts of their service life in zero volt conditions.

You know, when the device is turned off.

Are you sure about that, when voltage goes down current increases. That may or may not cause flash over in the cap "repeatedly running" at the lower " could create heat and breakdown of the electrolyte ? It's not low voltage like 12vdc Higher Voltages do strange things.

What do I know ? You guy's know more about it than me. Just seems logical if I consider it as an inductor or sometbing.

 

[Sceadwian]

Brownout, can you quote your sources for that piece of information? I've never found so much as a single source of a scientific reason for such a statement.

https://www.electro-tech-online.com/custompdfs/2011/04/PDF_GeneralTechnicalInformation.pdf

Provides hordes of detail about electrolytic capacitors how the oxide layer is formed and their operating voltage and it quiet clearly states that they're usable down to zero volts, and that even a reverse voltage of .6 volts (a diode drops worth) is permissible.

How does operating at a lower voltage produce more current? That logic is exactly backwards from real life, unless you're talking about at the same power level which has nothing to do with the voltage the cap is capable of just it's total power handling and ripple current abilities which are separate metrics from the voltage that the oxide layer thickness makes safe.

 

[tcmtech]

Are you sure about that, when voltage goes down current increases. That may or may not cause flash over in the cap "repeatedly running" at the lower " could create heat and breakdown of the electrolyte ? It's not low voltage like 12vdc Higher Voltages do strange things.

Huh? Wouldn't the internal ESR tend to cause a decrease in current as the circuit voltage drops?

In many circuits electrolytic capacitors work with widely varying voltages anywhere from zero when the circuit is off and all the capacitors are bled down all the way up to the circuits maximum peak operating voltage.

My experience with electrolytic is that if they run warm from high ripple voltages using s higher rated voltage capacitor helps compensate by having a larger mass and an inherently larger ripple voltage tolerance.

For example 10% ripple on a 25 volt cap only gives me a 2.5 volt tolerance while a 10% ripple on a 100 volt cap gives me 10 volts tolerance despite the capacitance values being equal which means that if I had 3 volts of ripple current on a 12 volt circuit the 25 volt capacitor would run warmer than what a 100 volt capacitor of the same capacitance would when used in the same circuit assuming they has equal ESR numbers.

Thats my understanding anyway.

 

[olly_k]

Interesting replies folks. I had also heard that electros are better off when used within a certain tolerance of their rated voltage but would like to see a good explanation. It is interesting that electros don't often have a very high shelf life, and that electrolytics that are properly rated and permanently connected to a supply suffer less than those who's demand varies but again I have never heard this from a definitive source! Maybe that depends on surge currents etc...

 

[carbonzit]

I had also often heard/read, back in the old days (1960s onward) that electrolytic capacitors needed to be operated close to their rated voltage in order to "form" correctly. Not knowing all the gory details of the electrochemistry of wet electrolytic caps, I never understood exactly why that was true.

This thread, like so many similar discussions, seems to be a fairly even blend of fact and folklore. It sure would be nice to be able to separate the two. Any pointers to papers on this matter from reliable sources (not Wikipedia!!!) would be much appreciated.

Some things to be settled:

• Are electrolytics really much different today than they were 50 years ago? (Meaning are they different in ways that affect their operation below their rated voltage.)
  
• What are the actual industry standard recommendations concerning electrolytic ratings and operation? (I'm thinking EIA, NEMA, other official organizations.)
  
• What are the negative effects, if any, of running electrolytics below their rated voltage? how about far below?

Inquisitive readers would like to know ...

 

[killivolt]

Huh? Wouldn't the internal ESR tend to cause a decrease in current as the circuit voltage drops?

In many circuits electrolytic capacitors work with widely varying voltages anywhere from zero when the circuit is off and all the capacitors are bled down all the way up to the circuits maximum peak operating voltage.

My experience with electrolytic is that if they run warm from high ripple voltages using s higher rated voltage capacitor helps compensate by having a larger mass and an inherently larger ripple voltage tolerance.

For example 10% ripple on a 25 volt cap only gives me a 2.5 volt tolerance while a 10% ripple on a 100 volt cap gives me 10 volts tolerance despite the capacitance values being equal which means that if I had 3 volts of ripple current on a 12 volt circuit the 25 volt capacitor would run warmer than what a 100 volt capacitor of the same capacitance would when used in the same circuit assuming they has equal ESR numbers.

Thats my understanding anyway.

That's the part I don't understand. I figured that the load would dictate the current and that the cap would tend to carry the current through regardless. But, I guess the EM field strength would run hand in hand as the voltage dropped and so would current? Take a motor for instance it would increase in current as the voltage dropped and if our cap were in series it seem reasonable to think it would also have the current increasing as well ?

I still don't get it. I'll probably understand more later cause, I'll bet were not finished talking about it yet.

Edit: When I worked at the TV repair shop, I learned that an increase above the operating voltage was ok but, it was the uf that had to be maintained. It wasn't until this thread that I ever thought about a decrease in operating voltage. I'm glad it was brought up.

 

[tcmtech]

That's the part I don't understand. I figured that the load would dictate the current and that the cap would tend to carry the current through regardless. But, I guess the EM field strength would run hand in hand as the voltage dropped and so would current? Take a motor for instance it would increase in current as the voltage dropped and if our cap were in series it seem reasonable to think it would also have the current increasing as well ?

Okay I see where you are going now. Induction motors do show an increase in their running amps at a specific load if the line voltage drops, but there is a limit to how far that effect goes depending on a number of factors which are heavily influenced by the motors inherent design below that point if the frequency is kept constant the output torque and relating power drop off and the input amps follow along proportionally.

If you take a capacitor to be the electrical polar opposite of an inductor you will see that it exhibits a specific impedance at a specific frequency and thusly can be modeled like a simple resistor where if the voltage drop across it is lowered while the frequency stays constant then the current flow through it will also drop proportionally.

Mostly I think this electrolytic capacitor application theory is just getting over analyzed really.

 

[killivolt]

Okay I see where you are going now. Induction motors do show an increase in their running amps at a specific load if the line voltage drops, but there is a limit to how far that effect goes depending on a number of factors which are heavily influenced by the motors inherent design below that point if the frequency is kept constant the output torque and relating power drop off and the input amps follow along proportionally.

If you take a capacitor to be the electrical polar opposite of an inductor you will see that it exhibits a specific impedance at a specific frequency and thusly can be modeled like a simple resistor where if the voltage drop across it is lowered while the frequency stays constant then the current flow through it will also drop proportionally.

Mostly I think this electrolytic capacitor application theory is just getting over analyzed really.

I have had to go out and diagnose compressor failures and once it drops past 10% I know that it will lead to overheating. The protection circuit will not blow and so it continues to heat and slave the compressor into melt down. Yet, again i say. Until now I never contemplated the cap. I got further confused about HV as it was spoken of that you can get flash over and put that into my head. (Spinning me)

A capacitor has really became interesting to me now in the past it was just the part used to increase efficiency or power on start up. I alway's thought the transistor to be more interesting. Now, RC circuits have become one of my favorites. Schmitt triggers etc.

 

[ronv]

Looks like it is not as big a problem today as it was years ago.

**broken link removed**

 

[carbonzit]

Looks like it is not as big a problem today as it was years ago.

**broken link removed**

I b'leeves the URL you want is actually this one (you had 2 URLs jammed together there, buddy). "Mic Kaczmarczik's Sprague Forming Info Information" (reading it now as we speak).

By the way, here's the relevant part:

Here is a quote from appendix C of the 1990 Sprague electrolytic capacitor catalog.
"When aluminum electrolytic capacitors were first developed, deterioration 'on-the-shelf' was a major problem and frequent replacement of stock parts was necessary. Additionally, use of capacitors for extended periods at small percentages of rated voltage permitted the dielectric oxide film to 'deform', just as it would 'on-the-shelf'. Both problems were solved in the early 1950s with the introduction of high-purity aluminum foil. Oxide film stability was greatly enhanced and today aluminum electrolytics can be used after storage, and at any percentage of rated voltage, without loss of capacitor quality."

Straight from the horse's mouth, as it were.

 

[ronv]

Yep, Then the second one.

**broken link removed**

 

[MikeMl]

...
This thread, like so many similar discussions, seems to be a fairly even blend of fact and folklore. It sure would be nice to be able to separate the two. Any pointers to papers on this matter from reliable sources (not Wikipedia!!!) would be much appreciated.
...

Put your money where your mouth is: Link me a WikiPedia Topic that deals with electronics that you consider to be in error.

 

[misterT]

This thread, like so many similar discussions, seems to be a fairly even blend of fact and folklore. It sure would be nice to be able to separate the two. Any pointers to papers on this matter from reliable sources (not Wikipedia!!!) would be much appreciated.

IEEE Xplore has thousands of papers on electrolytic capacitors. The most recent research seems to focus on estimating the condition and lifetime of a capacitor and on methods for obtaining the equivalent circuit of a capacitor. Looks like switch mode power supplies have a big role in motivating these studies.

If you don't have access to IEEE Xplore, but would like to read some article, I might be able to send it to you.. just one or two if you really find something that is really interesting to you.

 

[BrownOut]

Different manufactures have different recommendations. Some say you should not operate electrolytics lower than a certain percentage, while others say it's OK. In at least one manufacture's literature I looked at, the literature on their older products didn't mention a lower limit, but lit on their newer products did ( guess that high-purity aluminum didn't work out as good as they thought ). And I've worked in design houses that have a lower limit as part of their design rules. Given that electrolytics are among the most often failed components, I tend to want to adhere to the more restrictive guidelines.