Foamy Power Factors – Electrolytic Capacitors with KEMET

Well welcome to the macro fab engineering podcast where your guests Yvonne kilos

and Susanna and cosca from Kemet electronics

and we're your hosts

Parker, Dolman and Steven Craig.

This is episode 196.

Yvonne has over 19 years of electronic design and development experience in different industries ranging from aviation to industrial automation, and maker at heart. His technical experience surrounds the intrinsic requirements and detailed developments of circuits. In his downtime. He likes to develop escape room puzzles, sensors, and anything he can 3d print. Suzanna is an electrical engineer from Macedonia. She has been with cabinet for seven years as a product manager of electrolytic capacitors for three years. She loves solving problems and Sudoku, and is a fan of martial arts in the movie if man.

So I have Yvonne, have you ever developed a escape room that someone's actually been in? Well,

I've developed some of the puzzles that have other people been into. So it's always interesting, because a lot of people get to tell you, Hey, I like to do this, this, this and that. And at the end of the day, you are like, well, how are you gonna accomplish that? And so you kind of have to walk, walk them through what can be done and reasonably. Have you ever

done an electrical engineering puzzle? That's like every day right?

Eight to find Monday through

you must use this oscilloscope and multimeter to solve this problem.

Do not touch here, right? Yeah. The first thing bank? No, I haven't. Mostly because people that like to do the puzzles tend to be people that are very logical with respect to trying to put things together that are healing kind of thing. Not too much math, if you will. Gotcha.

So we have Yvonne and Susannah from Kemet today to talk about electrolytic capacitors. And so what is a who is Kemet the company for those that don't know? So here

at Cayman we are a well known for a being a capacitor company. High quality capacitors we have customers come to us certainly at trade shows and say we love your caps high quality you know when we can go to your guys are our go to guys. But certainly our portfolio has changed drastically from just capacitors, we have some really cool products, things including from wireless charging products like ferrite tiles, to vibration sensors, pyroelectric infrared sensors. And in really what gets us excited nowadays are magnetics inductors power inductors and chokes. So, so you know from listening to your podcast, we probably are very well rounded up in order to help out with some of those. So you want to pick a component x initiative type of thing. We have some knowledge with some of those parts in that kind of give us a good segue really on to in something I kind of wanted to touch on. Right now we're transmitting here from our Kemet headquarters in Fort Lauderdale, Florida. In we're about eight office away from our seal. So as we mentioned earlier, he just walked by. So in we have a lab. This is our cake lab. This is the Kevin application Intelligence Center. And in if you think about it, the company, Kemet is invested to get to people that are designing So our mission is to make design engineers job easier by being able to provide information on those subjects that you know, you need some some way of getting approachable type of data or information about our products. So that's who we are and where we're at.

Did Kim at begin as a capacitor company,

well came in. We're celebrating 100 years this year, actually. And so Kevin has his origins while they go, but we weren't always a capacitor company. We started like I said, 100 years when there were the tubes And he got a progress over with, you know, different aspects of the development of materials. So in, in our heart, we're a material company, we, we know a lot about different materials to make capacitors in now with our new acquisition as well that we have that knowledge on how to make inductors and magnetics.

Well, so we're mainly interested in talking today on electrolytic capacitors. And this actually sort of spawned from something that I can't remember if it was Parker, I but but we had mentioned on a podcast a few months ago that we were interested in having someone talk about electrolytic, capacitors and short of I know Parker doesn't like this word, but demystifying the secrets behind them. And really, actually, what's what's interesting is this sort of kind of, in a way spawned out of something that I was, I was dealing with myself, I know, I was designing a new power supply, and I was actually talking to someone who has limited knowledge and electronics. And we were designing this power supply together. And we were at the point where we were searching for electrolytic capacitors, we knew some of the aspects of the caps that we wanted. But he kept asking me, How do you know if this is good? How do you know if this is bad? How do you know where this goes? And what to do with it? And it kind of made me pause for a moment because I was like, I got to think more about that. Because like, how do I know that?

So Steven, how did you start searching your, your right electrolytic capacitor?

Well, so I knew there's a lot of this particular design was based off of reference designs, and a lot of gut feel from having done these before. So I knew that there was a good handful of options available. And mainly, it was just cruising over Mauser, no, you know, finding a handful of things that I knew, specifically like brands and particular aspects of, of capacitors, in order to fit the particular circuit that I was going for. But in terms of like, describing all of the information on a datasheet, to somebody who doesn't know much about electrical lytic, capacitors, they this particular person knew value and dimensions, that's what they were aware of. And they would they would, you know, pick a capacitor based off of those aspects. So I kind of want to dig deeper and go further into electric lytic capacitors and say, like, what makes them tick? And how do you pick? Well,

that's a very good question.

Like, any good design question, it all depends, right? So one of the most literal is let's think about a application side right in. So if you're thinking, for example, in let's say, something, everybody can relate to this, think about a, you're trying to harvest energy from a wind turbine. And in order to do so, you're picking up that that energy and you need to store it in, in, in, so you're going to install it into some batteries, and then you're going to want to use that energy later on, you're going to have to have while you're trying to get, say, for example, you're going to move a motor, for example, you're going to have to have those components in there, those electrolytic capacitors that are going to be able to handle the current. And in that current is going to be driven by what the load is going to do to you. Right. And so what what I'm trying I guess I'm trying to boil down to here is that a while a capacitor is a voltage device, right? We all know, like, what it does the voltage and if we can go one level deeper, we should. But at the end of the day, the current is one of the limiting factors that describes what your capacitor should have, should it be limited? That's kind of does that make sense? Do you have anything to add?

Well, yes. When choosing the electrolytic capacitor, you start of course, how much space you get. So, you had the right start with the dimensions the voltage right and what normally for the electrolytic capacitors is required this what is their life expectancy, how do they handle the How much do people can they can they handle so to get to the right one, we need to consider for design requirements are What would be the design of the application itself? Like the what is the ambient temperature surrounding, because all these drivers to get to the life expectancy of capacitor will, will drive how much heat it will get the insight in it. And this is driving the repo current. And the repo current gets from the ESR. Keep the ESR. So, all of these matters for electrolytic capacitor design.

So I've got a application that's probably a lot of engineers would come across is that say we have a USB power supply. So it's getting five volts at, you know, to say 250 milliamps, and we're running that through like a 3.3 volt LBO. And we need some electrolytic capacitors to help smooth that out. So what would we look at at specking? out some electrolytic capacitors to help that power supply do its job of making sure that 3.3 volts is as smooth as possible?

So it's got to do all with current, right? So So you mentioned right now you have a five volt and you're trying to get 3.3 volts out. But you didn't specify how much current and obviously, granted, one, one of the aspects of the current is not the nominal current, but that those currents, those transients, those ripple current. And so let me Let me boil down a little bit, because I think one of the things that Suzanna said, and he goes back to a lot of people throw some terms around without giving the proper definition before going into the whole details of what a capacitor is. So let me reel it back a little bit in this, this talk a little bit about you know how engineering one to one teaches everyone about an ideal capacitor, and they give you a paper you graduate, and you come out and they're like ESR, work into that section. Or you're just, you know, like film capacitors, anyone, you know, like, or even polymer, what are you talking about? Right? So it's not to say that, then you're talking about an LTO. But a lot of people may we may be doing inductors, or things of that nature. And then you get into inductors chokes in your like, manganese sink in the core or nickel sink in that core core. What do I need a low loss, high loss core, what well, so you get a lot of information that you kind of thought you knew about, but you start learning as you come out into the world. So Susana, brought in a interesting point, right? She said that when you're picking your capacitor, you want to have a nodule environment temperature, you want to know, your current and your current grip holds. So that means if the load is trying to remove current out of the sun, from your voltage rail, the current has to come from somewhere in those capacitors that are close to that load is going to deliver that current in a ripple. So let's say you're trying to turn on and off, I don't know, the next biggest supercomputer Right? Or not to go too far a pie or Raspberry Pi, or an Arduino switching really fast video or something in those trends, since of current that is trying to pull out of the voltage rail is going to try to remove that current from a capacitor, or why is the voltage start sagging. So when we talk about the ability for a load to pull out current out of a capacitor, we're talking about ESR. So an ideal capacitor, the energy within the capacitor when you store it would not have any loss, you will be able to deliver that energy right away. But in the real world, we have ESR, which is the definition is equivalent series resistor. And it's really just a resistor. It's not really a resistor but is the effective non resistor right? It's effectively a resistor in series with the capacitor in that resistor, like any resistor would is going to heat up and that's when your you know engineering one on one classes come in, it heats up at a I squared R rate right. So, your current square times your ESR is going to give that power dissipation of of that component. So you start looking at okay, so how much is that power dissipation going to heat up my component? So then you go into the datasheet and you find the thermal resistance in, in when you go in there, you have that thermal resistance, that's going to be okay, now I have my power I squared R, or I squared times your ESR. And then I'm gonna multiply that from the thermal resistance that is found on the datasheet. In in that is going to give you a how much temperature you're going to heat up that component above ambient. And so now you're talking about Okay, is it going to be past this the rating of the component? Is it going to be you know, how long is my life going to be affected based on temperature? So,

yes, what, what I can just to make things much more simpler for all the engineers designers. Kemet has an online electro life calculator that is actually providing this information or calculation on life expectancy. Normally, if you know some of the these basic as you mentioned, basic requirements of dimension and voltage. And how much report current you're gonna apply to this capacitor? With this online tool, don't my calculator will you are able to see exactly what will be the exactly estimated sorry, the core temperature, the thermal resistance, the life expectancy. So all these specs are kind of making it easier. It's not just the datasheet. We tried to, you know, make it easier for everyone.

So, we've mentioned ripple current, quite a bit and ripple current, correct me if I'm wrong, but that's almost entirely dependent upon the design the designer is aware of or designs in a particular ripple current and then picks a capacitor that can withstand that ripple current correct.

So there are some limited limitations, right. So our Of

course, yes. So, the electrolytic capacitors, as we mentioned, they can withstand ripple current, we specified that in in our spec sheets. However, there are some design aspects, if you are using them, let's say less stressed out than what they are rated about, then the life can be you can extend the life of what is there in the specification?

That makes sense. So for example, if you see the electrolytic capacitors, you normally see a life in hours reading, right? And can you explain a little bit about the life reading it? So from overall design perspective, that is the rating, temperature,

rated temperature rated? Voltage and rated triple current, what we say?

So it's almost like if it's completely maxed out, so to speak, maxed out, yes.

And so that drain is very interesting, because once you I mean, unless you're actually going to be designing something, you know, 500 volts and 105 105 Celsius, and in nine amps ripple current, you're going to have a extension of your life of the capacitor by quite a bit of a factor

mean that we have the rule of time saying for every 10 degrees drop the life of the electrolytic capacitor doubles. So you can consider that as one one point. Another thing is, as for the voltage there is there is no need to direct the voltage with electrolytic capacitors. But if you do decide to do that, we recommend going up to 15% duration of the vote because you're gonna get additional Extra Extra Life hours. But if you delete it more than 50% it's just a waste of why you chose that her high voltage capacitor. It absolutely really you can use it that add the rated voltage

and so on on the temperature ratings that would be the capacitor at ambient or whatever the ambient is plus the power dissipation temperature from the ripple current Correct.

Normally we say the ambient temperature is the future.

So if you're running this cap at 105 degrees Celsius and it's running, you know, max out ripple current and max out voltage, it's going to be producing more heat.

Correct. So the core is capable to withstand additional temperature, let's say additional 10 degrees, five degrees C. Okay.

And that's what you were saying earlier about the core temperature, right? Correct. Yes, I actually have, I just went to Mauser and pulled up a random data sheet for I just went to their electrolytic section and clicked on anything, and found a Panasonic datasheet. And so they actually call out ripple current, as milliamps RMS for these particular capacitors is ripple current always measured as RMS and will always be shown on a datasheet as RMS in ours,

yes. Correct. So and, you know, kind of talk about their datasheet. But from our perspective, it makes calculations easier, right? So the RMS part of it allows you to perform the I squared R calculation, for power dissipation, the definition by nature of an RMS is pretty much the DC equivalent of what you're going to be seen, right. And so as being DC is going to be dissipating that power. Now ripple is not the sees something that's going to be switching on and off. Therefore, that's why we take that RMS is to kind of average out that power in similar to a resistor, right? As you're powering as resistor on and off, you're not going to have that full current, but rather the ratio of the current in the heat dissipation from that.

Yeah, that makes sense. I mean, certainly, the the purpose of the ripple current is for power calculations with ESR. Correct. So you would think, because of that, RMS would make sense, for sure. So another thing, we've been talking a little bit about the life of a capacitor, what does it mean when the life of a capacitor is is over? Like when we say 2000 hours? Like, what happens at 2000 hours? Obviously, something magically doesn't just change. But what does that mean? Okay, so, capacitor, supernovas, yeah, it just explodes. What does it mean when the life is over.

So normally, we have several types of end of life. definitions, we can have, let's say, mechanical end of life, when the capacitor is actually having split sleeve or some of the vents start operating, we can have a catastrophic failure, right? That would be open or short. And from electrical parameters standpoint, that we normally state capacitance change the delta C, for, we have it specified for depending on the voltage ranges, either 10 15% or 20%, capacitance change, that would be the and then we have the ESR change. If it didn't, if it goes above what we specified, so we specified three times the initial value, then that's end of life. Also the leakage current if it goes out of spec, and as well, the impedance.

So it's interesting, we were talking about this, this afternoon a little bit, we picked up one of our data sheets in our effort to improve how to read some of these data sheets. And we looked at a couple of things with respect to life expectancy and how we define it. In I think, is going to make a great blog post on our site and perhaps a video on how to explain this a little bit. With respect to like, like Suzanna mentioned ESR, the ESR changes is one failure mechanism. In we specified the ESR within the datasheet. But does the ESR after which, if you go and have a higher DSR than that the capacitor is no longer read it to what the original state was or when you got it. So yes, I was one of those things in as in also, as you mentioned, and perhaps I'll let Suzanna discuss a little bit the internal or what a capacitor looks like inside but losing that capacitance value is because over the lifetime, the temperature the heat is going to be consuming the part of the cup As if that gives you capacitance. Okay, so

from a construction point of view, I have one bounding element that I can start speaking of from. So what we do is we get a foil that has been slated,

we'll need to we'll need to take a picture this we've got up on the video here, the actual guts, I've seen the inside of electrolytic before but usually because I connect them backwards and say yeah, they showed it showed me its guts in a way that I wasn't expecting. So

it almost looks like you could make it home. But believe me, you better later explain it better what that looks like inside.

So you have the foil and then wind it in between with paper, I'm having taps welded on on the amp foil and then those steps are being welded on the on on the deck itself, which is sexually making the connection terminals, two terminals right. And as for as for the capacitors change that we mentioned. So this boundary element is being soaked into the electrolyte, which is make making you know the case this conductivity for the capacitor and with time it dries up. And that is creating the the capacitance to change.

So it's the dielectric inside there changes which changes the capacitors level. Correct

is that the main factor of what causes a a, an electrolytic to age is it just the evaporation or the I guess the electrolytic being consumed.

The evaporation of the electrode electrolyte is caused by the ripple current that is going through it the stress of the voltage. So this is over time it's it's it just vaporizing and going through, let's say four screw terminals, we have the vent on the deck itself where the terminals are. Or let's say on snapping capacitors, we have a vent that is at the site of the can or the bottom.

So one of the things to think about like electrolytic capacitors and in maybe if we go back to some of the basics, a capacitor being two plates separated by a dielectric to electrolytic that dielectric on the electronics and electrolyte in in so when you build it you build it as if you imagine a I was gonna say a toilet paper but

cinnamon roll,

cinnamon roll that's a bear endure coiling that up inside the the package and you dip it into that electrolyte now, one of the things when you're doing that current generate in there that rippled current in the heat that we're talking about. With respect to ESR. One of the gas that get released within a electrolytic is actually hydrogen. And in as you consume or your device is being used, and it's getting consumed, I guess in a way that electrolyte is what is going to start losing weight if you will from from the component itself. And your case as you lose that electrolyte that dielectric. The dielectric is everything within a capacitor. And based on that dielectric, you're going to lose the storage capacity of that capacitor

just as a random tangent. Can you refill a capacitor? No.

Or is that just what big candidate wants I'm imagining like capacitor in like where they like inject like steaks for grilling, like Rose big syringes.

Yeah. Oh, yeah.

I'm assuming that electrolyte is not the same as those sports drinks so therefore, you're not going

to filled with Gatorade.

I I think we did actually have a conversation about this on the Slack channel. It's like what what could you put back in a capacitor to refill it? And sweat was one of them. And Gatorade was another just trying to figure out so but but I guess the Yeah, the easy answer is no, you can't do that right?

Not advisor.

So, so the the internal structure and the actual design and how the capacitor is constructed is actually how things like ESR come about, right? Because ESR is the actual plates themselves have resistance in the weld tabs and things have resistance also, right. Like we were saying there's not an actual resistor inside.

Correct. So all all these that you mentioned, are one of the main contributors for for the ESR. That taps the paper that the foil the electrolyte, so everything is contributing for that. The ESR.

So do do electrolytic, do they heat up generally fairly even? Or do they have hot spots? That you have to worry about?

What what I've learned so far is that it's normally the hotspot is somewhere to the 1/3 and the capacitor size

to 1/3 of the way up. Yes.

That's, that's interesting. I'm wondering, I wonder what mechanism causes it to be about that?

Well, not until one thing I've kind of wanted to mention in something we forgot about is, you'll be amazed how much a little bit of cooling is going to increase the life of your components. And in specific electrolytic. And removing that heat from the component obviously is going to, you know is going to release is going to use up less of that capacitor life, if you will. So just a little bit of air is amazing how much it does. 12. Anyway,

very helpful. Yes.

Do. Okay, so another another? I guess tangent? I feel like we've got a lot of tangents going here. But I like it. Do do capacitors change their dimensions throughout their life? In other words, do they swell? Or are they pretty stable? I mean, I know I've seen bad capacitors and they have the puffy top on them. And, and you know, that's what it's a bad ambassador, but does a good capacitor change its dimensions,

it should not. At least electrolysis should not should not change the any of the damages. As you mentioned all those paths, I guess that would be the you know, the disk that will be swollen or even the sleep or so that that will come to its end very soon.

So if you notice, in the construction of a lot of electrolytic capacitors, and you may ask yourself that for is getting a little bit of cut out on top like a square or like a star shape. Not really a color but like a V scoring on top of them, which is acts as a valve if you will for pressure release. So that electrolytics as I mentioned, they do tend to release hydrogen. And so one of the things that that allows to is for easy scape have that pressure built up within the component.

I think we have one paper already written by our Fe sem about the hydrogen generated in the electrolytic capacitors. That's a good point to have it shared.

And I'll take a picture of this, but you're talking about the plus sign the plus sign. That's that. So that is purely for if it does, catastrophic, we fail. It doesn't catch off. You feel really bad. Right? It does need it has a mechanism to release early I guess, pressure valve? Yeah.

Is that because the electrolyte boils?

Um, I'm not sure if it boils, but it's more because of the guests insight that the pressure and then it gets released in that way. Normally, when it's working in under conditions, there is again this pressure building up but it's it's going slowly. But if if it gets overstressed then it

will lower gears when we talk about being able to size your capacitor correctly for the right ripple current. Right. So let's say you do have your application and you say oh our, our application can only handle two amps of ripple current. But then you go and connect something crazy and start using 20 amps of ripple current over is certain amount of period of time. So that's when you start seeing these type of issues when you're over stressing a component in the way that it shouldn't be. So that life starts being affected by by the ripple current. Again, that ripple current, obviously, like I mentioned, it's kind of, we've talked about it quite a bit, but I squared R is definitely something to watch out for, if if anything. So it is not only that, but it only if you think about it is not only affects electoral politics, it also is going to affect any of the other dielectric. So film, they have exactly the same mechanism. In a lower, obviously a lower scale. mlccs also have that. So that ESR part of the overall construction of the capacitor is going to generate heat that's going to, you know, heat doesn't electronics doesn't work well together.

So ripple current in ESR, are in a way inherently linked, right, certainly. So what other factors exist in a data sheet that we would need to know about?

That's a good question, because it only until you need something is well, oh, let me check the data sheet that have in there. But one of the things that a lot of people like to go in is very important is the impedance. Impedance is one of those tools that if in, we probably need to come back again to electronics one to one right. And in in discuss what is in tenants in an impedance of a capacitor, at its core value is going to be, it's going to have a resistive value, which we've been talking about is the ESR part of it. And then you're going to add that reactive value, which is going to be the reactive the capacitive reactive and the inductance reactive value for in those are going to be some values that are going to change based on frequency. Okay, so if you go back way back to your early classes, the impedance of a capacitor is one over two pi frequency times cap. And so you can see how as you're increasing frequency, being in the denominator, your impedance is going to be lower. And if you think about the capacitor as a whole, you can drive a line sloping down with your impedance on the left and on the y axis and in on the x axis, you have your frequency that sloped down of impedance or reactance, capacitive reactance is going to slow down. But as you get higher and higher in frequency, you're going to see that the inductive reactance value of the capacitor is going to start increasing. And if you go back to your overall equations, the reactive inductance is equal to two pi times the frequency times the inductance. And so as you go up in frequency that's going to go and show you a slope, that's going to go up. And so as you see, at one point in frequency where those two lines cross, you're going to have a point. And that's called actually the self self resonance frequency. In that point is where your capacitor stops being a cap. And that's for pretty much all capacitors. So you're trying to get to a frequency value that you're that's not work close to that self resonant value, right. And a lot of people go to the datasheet to go look for that up, or usually is somewhat difficult to display. If you imagine all the again, if you remember that equation, one over two pi frequent two to pi frequency cap. If we're going to put every single impedance value for every single capacitor and every single frequency, you're going to have this mess of our graph that you're going to try to understand. So in order to do that, we actually have developed a really cool tool, our casing tool online, is similar. In with that simulator, you can go on actually simulate a couple of different things. You can go check out impedance and ESR. As I said, the impedance, the ESR is always going to be there. And on top of that, the ESR also changes with frequency, but you're also going to be able to find things like, for example, I mentioned the SR, capacitance drop, if there's any, in, you're also able to see some of your S one, one and S one two values that you want to have for overall modeling of a component, if you want to do that, in also in there, you're going to see the model of a capacitor, what it looks like with resistors. And in how the capacitors get affected. For example, we talked about insulation, right? Whenever you think of a capacitor is just think of two plates. Well, there is actually a resistor that goes across those plates, which is going to be your insulation factor. Two, is your design going to be going over that insulation? Right? What what do we need to know about your design in order to make sure that that those factors within the datasheet get are not violated? So there's quite a bit of things. So without really going into much detail on specific datasheet, you have a whole bunch of things. Now. I know for a fact, a lot of people look at a

characteristic of a capacitor and many wonders if they need to use it. And that's always been the case of the dissipation factor. And I don't want to go too deep into that. Unless you want me to, well,

yeah, first, what is dissipation factor?

Oh, we're gonna go into alright. grab, grab your cup of coffee. And since we already talked about ESR, then may as well go into and we talked about impedance may as well go into the patient vector. Well, given that you guys brew, I think it's even the new brew some

beer. But both Parker and I do yeah.

Oh, there you go. So you'll, you'll find this correlation.

I'm excited electrolytic, capacitors and beer brew, and you're speaking my language. Now.

The dissipation factor, the way I look at it, and you'll probably find a correlation to something else as well is, if you think of beer, right, so three fourths of that beer is going to be my, my good beer that I'm going to be able to drink. And one four of that is going to be the formatter that some people like. Similar in the dissipation factor, that ratio of foam to actual beer is your dissipation factor. And probably you heard this similar analogy with a power factor. So they're a little bit correlated, and it's a number that gives you the ratio of how well the energy that you store can be released. So it gives you that known ideal character value for that non ideal capacitor, right. And the way you define that is the ESR over the impedance of the capacitor, so that's, that's your into as as, as your ESR is higher, your quality, that the factor is going to be higher. So you want that number to be lower to get it as close as possible to your ideal situation of a cap, which you'll never find, right? So that's, that's one way to explain now. It's also called tangent Delta, or loss tangent. And, in so I'm gonna, we all have heard from our circuit basics, that that in a capacitor, the current leads the voltage, right? But we never like a lot of people like alright, they think about it, they, but they don't really like Okay, put it in a quiz, you're, in reality, if you think about this, it's called that is because if you try to put a voltage across a cap, that cap is going to look as a short leg as a resistor that has a very low value, and that current is going to go straight to to ground or whatever you grab, it is going to chunk that if you will. And as as the capacitor gets building up that gets being charged, that current start tapering off and the voltage keeps rising, right. So that that's why that current the the amount of current that comes out is higher and is leading the voltage because as I showed, you're not going to have any voltage, but the current is going to be Big, therefore, the current is laying the voltage. Now it tells they always tell us the current is being led by 90 degrees. Right. So if you plot that in a sine wave, if you apply a sine wave across that capacitor, the other 90 degree shift, the current is going to be heating that first and then the voltage right. Now, the dissipation factor and why I mentioned in whole days is because the dissipation factor is the how that current doesn't heat that 90 degree due to the impurities and the non ideal parts of the cup, and is going to be less than 90 degrees in the amount of angle from 90 degrees, where the current is going to hit is your delta. And when you take the 10 Delta and the tangent of that delta, that's going to give you your dissipation factor. So, a lot of cool math, a lot of cool in that happens to be equal to ESR over the impedance, right? So, very cool stuff. But like we talked about, some people may want to use the impedance, for some switching frequency, make sure that you don't go into self resonance. So people want to use the ESR make sure it doesn't heat up amazing. But the dissipation factor is not everybody uses their special cases. Yes, there's special people that like to have it. But definitely, it's something that not everybody uses. But it's cool to know about and we use it certainly for quality control. It allows you to know how good a cap is, right? But apart from that, is one of those things in the datasheet that you look at it and say and wonder, Am I okay with this part?

So, you were talking about the quality of the cap? Is that if it's if that 10, delta is closer to zero, so it's closer to 90?

Correct? Okay. 99.

Better, right? You're into our values of dissipation factor.

Correct. Except, so here's where you probably heard of the quality factor or the cue of a cat. And so when those numbers we can so small, it's kind of hard to say, Oh, he's got a point or five, right? So we do that one over the dissipation factor, and we get Q. And so that that way tells us we were able to use bigger numbers instead of saying point 0000.

For anyone, for anyone who's listening right now, that is a student in Electrical Engineering. I think I think you said it perfectly. earlier. Yeah, you get a piece of paper, and then you start learning everything. Like that piece of paper just means that you survived four years of college. And now you get to learn the real stuff.

I'm telling you, certainly.

So the dissipation factor is is kind of connects the dots between different aspects of the capacitors, right? Like you were saying it's, well, it's, it's how off the the actual the current is from a pure reactive load, right?

Correct. Yeah, how often is from perfect,

but but the reason, but the reason that it is off is due to the the, the capacitor not being perfect. So it's due to the ESR. And due to other aspects of the capacitor write DSL to

write in only that. So if you think about it, there's an based on the dielectric, if you think about it, that ESR in the impedance is something that is tightly coupled to the frequency. So the dissipation factor has to be tied to a frequency. Therefore, for example, you'll notice in our data sheets, we test the dissipation factor 120 hertz, in one volt RMS, in only that, it also could change because of temperature, right? Like Kappa, we talked about capacitor and changes in temperature and all this. Different dielectrics have different temperatures in so we tested at 25. C. So that dissipation factor is certainly some number that can move based on your frequency and temperature. So, yeah, if you're using you know, what value you need,

right, and I think I think really what, what a lot of this, for me personally is kind of connecting the dots is you have all of these factors on the datasheet. And there's a there's a good chance that you you need to really care about one or two of them, and maybe a lot less about the others. But you have to know when to care about those right or your design dictates which ones are the most important. Okay.

So I think earlier you were talking About the dissipation factor is something that not a lot of people have to worry about. When what kind of application would there be, you'd have to worry about this thing.

Because you asked me that. So, there are in we do market, some mlccs, in when you care about this is when when the amount of storage on your cap is needed to deliver different things, right. So, for example, we do sell some high Q mlccs, or SI BR series. And in those who are target, for example, to people in RF, when, when, when we're doing power in, let's say, for example, resonating circuits, some of the resonators that you're wanting to have that amount of power available to you and move away from or you have to consider that based on you know, you want to change the ESR or change on these things. That dissipation factor is going to be critical to them. In and if you think about in you look at resonators their tune into when resonator in a nutshell is you put a square wave in the one end and you get a sine wave in the other and for most of time, right. And so that that resonating value is going to need some current that goes along with it. And you want to be able to maximize that current, make sure that doesn't to those values in there, those those designs where they're important has to be in an essence like resonators. And in RF application certainly

even you mentioned the capacitor as storage, but it reminded me one fact about the little electrolytic capacitors, when you actually store the electrolytic capacitors for a longer time, things that designers need to have in mind is that they build up the leakage current will build up over time just sitting on the shelf. So, normally we specify and when when they are needed to be put in use, we we suggest that they are slowly increasing the voltage right. So to ramp up slowly the volt voltage and have it having the voltage applied for for a certain period of time. This is this is something that is pretty much recommended. And also another fun fact is that they they can recharge by themselves. So because of that the dielectric absorption, they they can do the voltage inside them. Approximately 15%.

Just over there.

Yes. That's always fun. So,

so moving on the big ones, right?

The screw terminals, so those big ones that in my big I'm going to what can I compare them to?

It's like a Coke Can I sprinkled? Yeah.

A Pringles can. In that, that's going to, if you leave it sitting around, the cap is going to gently started generating some voltage across. So actually, to store them, we toured them out. So that we you know, when we go to trade shows and things like that, for those components that are stored and then shipped out, so that people that just don't handle them, because those are heavily read like a high voltage trade. And so if you think of a component that's rated, what's one, what's the high voltage, right?

Like, now we have 550 miles from the big screw terminals,

okay. And, and so we have, you know, this big up, you know, I probably assumed to answer 700 volt electrolytic. If you think about 15% of that electrolytic being charged top, and you go touch it, ouch. Right. And so, that's one of those things. And but yeah,

so I'm gonna miss this. What is the mechanics of getting that charge? It's not just absorbing it by static, it's what's the mechanism of it? Because Because that's the thing is it's the voltage, its potential. So work was put into the capacitor. So where did they get that energy from? to get that 15% charge?

And that is a very good question. And I think we are gonna write some blog about that, because

that's an excellent answer. Yeah. All right. A blog about that. Yeah. Okay, so how about these new polymer electrolytic capacitors, what's what's so special about those

so that's, that's a, I think we can have another podcast about that

we write a blog about it I mean,

two blocks above

of course. So, normally we talk about the electrolytic capacitors, the the web technology, but then the polymers comes in and say, Well, we are much better than you in terms of life, when you you have them operating a little bit D rated than the rated temperature. So, if you compare, I mentioned earlier about the wet electromagnetics the life expectancy, the life is as per the rule of thumb, if you decrease 10 degrees temperature, the life doubles. Well, with the polymers, if you decrease 20 degrees C, the life is 10 times more. So this is coming up from from because there is less heat generating during these operations compared to the standard telekinetics. So

they just have a lower ESR. From the get go, okay,

they have much, much more ESR. It's so the ripple current capability is much higher. Also, there is no electrolyte leakage. Potential also extends for them that there is no need of voltage derating. And I must mention that currently, we are launching extension of this new SMD series. So it's not extension, it's new s&p series that is going up to 125 degrees C, which is pretty cool.

So I like these the polymers in particular, they there is a great technology, in a sense that, as you mentioned, Parker, you hit it on the ESR value drops quite a bit, and allows those higher ripple currents to occur. And in so in my perspective that applies. Now, one of the things that there's always drawbacks in one of the things with electrical with Polymer, aluminum polymer is that you have a little bit of a higher leakage current, is that correct?

Yes. So one of the things we need to be careful if your application is leakage current sensitive, the polymer might not be the best choice.

It I'm also just noticing, and maybe this isn't necessarily true for Kemet, but polymer caps are expensive. They're, they're quite a bit more expensive than electrodes.

Right? In, you know, obviously, the benefits of longer life is something that is very attractive, right. So, as we talked about the as it allows you to do 20 degree drop in your rate, it gives you 10 times the, the lifetime, right? Yeah, I mean,

that's one of the things but then when you consider if you pick and choose same dimensions, voltage that we discussed, we got to do this experiment, right, right to compare the electrolytic with electrolytic. And the polymer. The because of this ESR difference, how many capacitor how many polymer capacitors you would need to use? Versus with with when you apply the same, let's say ripple current to look to them. So that will make the difference? Maybe you would be using let's say, I'll just drop a number, like 20 wet and then you just gonna need for polymers that would be so the trade off? Sure doesn't necessarily mean yeah, you you need one and then compare it that one with the one Yeah, it will be more expensive the polymer but there is a trade off

in one of the things to think about if you're thinking of like say a switching regulator, right? Some of that ripple voltage when you're doing your smoothie now out of your of your voltage output, that ripple current is going to in both the current and the voltage, that ESR is going to come Was that report voltage. And it's got to do with the fact that the ripple current as you're switching on and off that the, the inductance is that current is going to go fluctuating over the capacitor that ESR is going to generate some of that ripple voltage you're going to see on your voltage output. And so as you're going to do try to minimize that ripple by lowering your ESR. And so to do that, you can add different types of capacitors in a bank. So let's say you put a whole bunch of aluminum electrolytic capacitors together in order to reduce the ripple voltage, or you can use one of the polymers and you're going to be reusing that ripple voltage, obviously. Now one of the things that kind of came to mind when I was talking about this is our casing tool actually allows you to put together different types of caps, and be able to see how that impedance gets dropped in a, you know, in, in a waveform type of output. So, so let's say you, you're like, oh, I need this type of impedance over a frequency range, then you start adding, let's say, your, your, your electrolytic, and then you add your mlccs, then you put some polymers in there, and then you have a full why ban frequency that you're targeting during that filter. And so something interesting to look at, when in I don't know, if I mentioned our casing tool, this casing that came at that come and so go out in there and play with it, it's really fun just to see what will happen or how those different parameters going to affect the the values of your impedance.

That's actually it was the, the example I was gonna give Steven is usually when you're designing a switching regulator, you need your capacitor output capacitors usually have to be kind of low ESR. Or if not, they tend to heat up a lot and then explode. The first couple switching regulators I tried to design myself, they exploded. And, and so you generally use last ceramic capacitors for it. But the problem of ceramic capacitors is you typically other low values, and you need to get into, you know, the couple of these couple of 100 micro farad range A lot of times, and so you either go okay, sorry, I can't put enough ceramic capacitors because it just takes up too much board real estate. So I'm gonna go, tantalum, tantalum 's are really expensive capacitors. And so but electrolytic, you could put electrolytic in there. But typically they had ESR is that were too high, those would heat up and explode. So I haven't played a lot with polymers. But polymers might be that middle ground where you can get away with a couple ceramics and then a polymer instead of having to go with ceramics and then a tantalum. I don't know, get it, I don't want to guess

here's something that I'd like you to kind of, you know, in because we cannot get involved a little bit with our discussion on Polymer alone. So when we talked about Polymer, we're talking about aluminum polymer. So great. The construction is a little bit different than for example, our tantalum polymer, which is different than our tantalum Eminonu. Which is so if you think about it, those yellow that they when people talk about tantalum, they refer to those yellow capacitors, which are tantalum M and O two, into our tantalum polymer is another technology within our portfolio that's capable of having a high density and take advantage of these polymer characteristics in a small package. And so you have a tantalum as a as a material this is really rich and dense. I don't think there's anything else that can contrast the the capabilities of the polymer to deliver energy. And being polymer makes him really safe. In so that you can contrast a little bit with the aluminum polymer, which is also a low ESR component, low use or value. So different technologies to pick from different designs criterias voltage levels that you need to be able to support your design based on what you need. So yeah.

All right. So we have a couple of questions that were actually generated from our Slack channel. And you mind just answering one or two of those? Oh, yeah, we, we kind of just put it out there this week saying, Hey, we're gonna have some, some lunch. little capacitor people on, send us your questions. And we got a whole bunch of them, which is fantastic. And we already covered most of it. Yeah, we did. I've been I've been knocking them off the list. So yeah, we'll do one or two of them here. So this one, I'm actually really interested on this also. So how the difference between leaving a capacitor on in other words, it's in its, its in its steady state mode of receiving current and, and given current versus turning a device off and having it sustain that initial inrush is there an impact on the life of the capacitor one way or the other like, multiple inrush currents or just forever on?

Well, inrush currents we we've done, pretty much, we have some testings that we need to go through, and these are have to say, the screw terminals and and the snap ins, these are capacitors that can handle pretty high inrush currents. So it really depends, you know, if you're using it. And inside this specified limits that we mentioned all the time, they can, they will live up to what is specified. But if it's being used short, inrush currents and having on and off for a shorter period of time, then it's still you're gonna get the best out of the capacitor, we really have some high, high height this thing, values that we need to meet. So for IC certifications,

if you think about it, current obviously, as you mentioned, is the heating mechanism for that capacitor. Right. And so definitely, how you look at that current and I'm not too sure, you know, it's kind of hard to say, you're going to have full current on that cap. Does your hard work and raise your voltage has to drop or something. But but certainly that the current is the one of those characteristics that like we talked about that ESR value is going to generate a hotspot and it's going to be the end of the day? What limits the ability of your capacitor to last certain periods of time?

Yeah, I would also imagine it would also depend on the application because if you're running that capacitor Full Tilt all the time when your devices on it's a 10,000 hour rated cap, you got a year and a couple of months of that device working whereas when hours Yeah 1000 hours or like you turn it off when you're not using it right.

I guess the point was the inrush current can be significantly higher than its ripple current right. So is that does that cause more damage to the capacitor than running it at its ripple current

normally is as I said, these screw terminals and snap ins that we talk about they can handle very high revoker inrush current so they they are not because those in inrush currents are for very short period of time and they cannot heat up quick enough the cord that is actually affecting getting to the life shortage, right? It makes sense.

And it probably also goes back to if like the caps are sitting for a long time slowly ramping up the voltage to prevent them from losing too much of their you know electrolytic juice. Yeah.

Okay, another quick one about the actual screw screw terminals. One of our listeners was asking about are the actual screws that go into the screw terminal are they intended to also carry current or are they just purely there to hold down whatever terminals you have? And also should you do your best to match the material of the screw to the screw terminals material itself?

So if I understood well the question is the screw terminals are passing current,

the actual screw itself is the screw intended to pass current along with it or is that not?

So the connection? The question is if this crew is the one that's going inside the the terminal is this crew meant to be made. So if you were to To make it out of plastic, we're expecting this crew to be the part that that is current, or is the terminal connection

terminal, the terminal connection is the one that's passing through the current. So it has to be connected to something that is passing through.

So you could technically use like a nylon screw, if you can get the torque the nylon, because there's a torque rating for them, right for the terminals. So if you use a nylon screw, it would still function correctly.

We have some accessories for the screw terminals, and they are nylon screws to Yeah, okay.

So there you go. Yeah. So yeah,

I guess I guess you're not banking on the fact that the screw will also pass current, the terminal is rated at the full current, which most of the time, those terminals are pretty hefty and beefy. But then also, the Yeah, the actual terminals are so there. Are they aluminum? I think so. Right. So should the screw if it's, you know, if it is a metal, should it also be aluminum?

Um, I'm not sure about that side of the part. But I guess

I'll have to because yeah, the similar dissimilar metals that dissimilar metal that you know, that makes another perfect Blogspot. I don't think I have think about that. And so we'd certainly put that down. And it'd be interesting to find out more, so we'll write something up.

Yeah, I guess also, if you're if you're thinking about writing a blog post on that, if you're doing a Buspar or something, should that also match the material of the terminals? I would think so write anything that makes contact with it, you would want to be similar? Metals?

Steven, we've got the redesigned to super simple powers. Yeah,

you're right, because I had copper bars going right across it? Honestly, I don't I don't know the metallurgy enough to know if that's a problem. Well, fantastic. Parker, do you have a you have anything you want to add to that?

I don't think so. I think I actually I have I have one more question. And it deals with electronics as well. And this might be getting a little out of the weeds given a current topics, but signals through electrolytic capacitors. Because usually, when you're dealing with like, when you send in signals, like RF signals or audio signals, you use a polarized capacitors a lot of times electrolytics to be a DC blocker. Why is that a thing? Stephen might have in that connection? Add to that question. I don't know. Like, why does a capacitor in series to a signal, block DC? Voltage, but allow the AC signal to go through?

Oh, the actual wire and capacitor? Let's turn go through?

No, no. Why does it allow the AC side of a signal go through but blocks dc side?

They put it in series? Yeah. Yeah. We need about two, three months worth of Maxwell equations to tell you that in a little bit. Because yeah, you're right, right. So if we put a cap in series, they are completely isolated both plates by a dielectric. And so how that current goes through? I mean, when you know that if you put a resistor in a loop, they're going to charge up that capsule current must be going through right. But certainly in order to answer that has to do with you know, the the electric field that generated within the capacitor dielectric and how that, you know, it's a full professor type, physics and materials, guys. Maxwell equation type of question that? I don't think I can answer.

So actually. So as sort of a sidebar to that, how about just okay, instead of capacitors in series or anything like that, most of the time you see electrolytic capacitors as bulk storage, or for smoothing purposes, you don't see you see film and ceramic and things like that more for actually passing signals through them. Why don't we use electrolytic to pass signals through? Like if you were doing coupling between amplifier stages and things like that you don't normally see that being done by an electrolytic. What's why not?

So electrolytic has this great advantage over a lot of the other components is one of them, obviously price, but the other one is its ability to hold a lot of charge. And so it's the question is, well, why don't we use that truck to go to work? Well, I I mean, yeah. But But overall, you kind of want to use the right tools for the right thing, right. So electrolytic tend to be both here. In they are, because of its construction. They, they're polarized. And so normally when you have a, let's say, a series, a cap that you're putting in series, as you mentioned earlier, you're going to be passing a see through, right. And so in essence, you can have a polarized capacitor. And so that's why film, in other words, is utilizing in those aspects. In in the, if you think of film, it doesn't have any polarization. And so it's great to make, you know, like simple power supplies and things of that nature. So I don't know if that answered the question, but it's kind of towards that end, right. Then similarly, you can use ml disease to pass small signals or smell AC in filter those out?

No, absolutely. I think I think that answered the question. I think the answer boils down to a lot of it's just not the right cap for that situation. That's not saying you couldn't do it. It's just, there's better options.

Correct. And again, the polarity, making sure that the polarity doesn't know reverse bias. So some caps do not like to be reverse bias whatsoever.

So if Steven, you have any more question, no,

I think that's good. We really appreciate you guys coming on and enlightening us on a lot of these topics. That was certainly Fantastic.

Thank you. And so where can our listeners find out more about Kemet and the tools that you're talking about today?

Well, certainly. So I wanted to, you know, give our gelato or cake lab. We do a lot of things here. We want to grow our community our feedback, our engagement to you guys can obviously go to engineering center.com forward slash cake, that's k Ay, ay Z. And I can't believe we didn't make enough jokes about cake because we have all the puns here. But But certainly, our cake lab is we are here to help. We're here to try to come up with videos that help different engineers with questions that maybe everybody has in nobody has answered to so we kind of want to make sure that we are part of that conversation. And if you guys want to reach us we again, that's engineering center.com. And then obviously we have our social media channels. We would love to hear from you from social media makes it easier faster. Our Twitter account we're Kemet capacitors, we also have an Instagram account at Canada electronics. And you can follow us on YouTube. Facebook, we were pretty active in there. And if you ask us will, will, will will answer. So go, you know, follow us and we'll try to get to all your questions as soon as we can.

And I will be looking forward to those blog posts that y'all been talking about. And with that, would y'all want to sign us out?

Certainly. So this was the micro macro engineering podcast. We were your guest, Yvonne Kidoz

and Susanna Costa.

And we're your host Parker Dolman Steven Craig.

Later everyone take it easy

Thank you. Yes, you our listener for downloading our show if you have a cool idea, project or topic. Let Stephen and I know Tweet us at Mack fab at Longhorn engineer or at analog E and G or email us at podcast at Mack fab.com. Also check out our Slack channel. If you're not subscribed to the podcast yet, click that subscribe button. That way you get the latest episode of right when it releases and please review us wherever you listen, as it helps the show stay visible and helps new listeners find us