Why Your 4.7µF Capacitor Becomes a 0.33µF Capacitor

[Carrington]

That software is free. I have already posted the link to such software on the forum here.

Please. Where are these links?
Thanks.

[alm]

A Google search quickly found this:
https://www.eevblog.com/forum/beginners/esr-meter/msg109145/#msg109145

[free_electron]

That is why those barium titanate dielectricc aps like x7r are a nono in filters or amplifiers that will be mounted in vinration rich environment !

Well you can make "acceptable" filters with NP0 or COG capacitors. X5R and X7R is suitable for power supply, bypassing and such applications. I really expect a superior technology to be available in mass quantities in the near future, as there is the demand for it. In fact we already have it in the 18-20 bit SAR ADCs and zero drift PGAs. But it is time to give something for the discrete guys.

Of course. That's what i said. NPO dont use BaTi as a dielectric so there is no reset behavior due to soldering. C0g is the jedec name for NPO. C0g is in the same line as x7r and y5v naming convention. Letter number letter

NP0 means negative positive zero. No temperature drift.

[Carrington]

A Google search quickly found this:
https://www.eevblog.com/forum/beginners/esr-meter/msg109145/#msg109145

alm thank you very much. 
It is the same software that I had. Excepting Murata's SimSurf. Instead I had used Murata Chip Capacitor Characteristics Data Library.
UltraCap also is interesting, although it is not the same type of software.

Woow this forum is great. 

[Poe]

Here is another one: a freshly soldered ceramic cap decays at a rate of 1% the first hour, 1% the next day, 1% the next month and a 1% the next 2 years (it's a 1:25 ratio... 24 hours a day , 24 days a month , 24 months : 2 years).

Heating the dielectric above 148 degree centrigrade resets the decay process !

So: you can't make frequency stable oscillators with these suckers ! Worse. If you got the frequency spot on , don't reheat the cap : it'll reset !

Because of the trend to replace expensive seasoned engineers with cheap green engineering students, our company accepted the return of >100k PCBAs due to "poor solder joints".  Fresh out of college myself, I discovered the actual cause was due to this 'aging' effect a month later while reading this tech paper:
http://www.johansondielectrics.com/technical-notes/product-training/basics-of-ceramic-chip-capacitors.html

Really wish companies still understood the value of mentor programs.  A single engineer with a dozen years experience would have immediately known their timing circuit should not have relied on a cheap capacitor... and that they could not hold calibration because their techs just got done reflowing the cap. 

[c4757p]

*I* know not to use a ceramic cap for timing and I have no industry experience! It's too bad that the companies don't have mentor programs, but it is equally sad that the students don't have the motivation to pursue extracurricular experience before they graduate.

[Kjelt]

It's too bad that the companies don't have mentor programs, but it is equally sad that the students don't have the motivation to pursue extracurricular experience before they graduate.

Sad that you blame the students while the real blame actually IMO lies with the colleges and universities that skimp on the education and society that pressures students to graduate asap. What is really shocking is the difference in educational institutions, a MscEE from a midlevel university should be equal to an MscEE from MIT, that this is obviously not the case is proof of the problem that a level of education can not really be measured in a degree.
Then there is the big difference between dry theory and the cold and harsh real world of electronics,  the world is moving fast and there is a lot to learn even for seniors.

The given problem of scrapping >100k PCB's by a certain company because they fired the (too expensive) seniors and let a (cheap) junior design and create the pcb's without rigorous testing or second opinion is just what they deserve, you get what you pay for and I just can't help  my ass off. I just hope that the responsible manager is still there and now has to inform his management about the loss.

[westfw]

the real blame actually IMO lies with the colleges and universities that skimp on the education and society that pressures students to graduate asap.
  :
Then there is the big difference between dry theory and the cold and harsh real world of electronics

Ok, I'm curious.  Did anyone with a BSEE or MSEE actually have a college class (or lab) that talked about the different characteristics of real-world capacitor types?  I don't recall any such thing; I'm not even sure whether they covered polarized vs non-polarized caps.  (too busy using calculus to figure out the capacitance of a plate capacitor with a non-uniform dielectric.  That was SO useful!)

[IanB]

Ok, I'm curious.  Did anyone with a BSEE or MSEE actually have a college class (or lab) that talked about the different characteristics of real-world capacitor types?  I don't recall any such thing; I'm not even sure whether they covered polarized vs non-polarized caps.  (too busy using calculus to figure out the capacitance of a plate capacitor with a non-uniform dielectric.  That was SO useful!)

I doubt it. This would be called "on the job learning".

It's a similar story in my field (chemical engineering). As a student you learn precious little about the practical operating characteristics and limitations of real equipment. (Some engineering schools are trying to correct the balance here, but it's far from being a universal philosophy.)

The thing is, it is easier to learn the practical stuff hands on at work and the underlying theory in college than the opposite. If you got taught the practical things in college and had to learn the theory on the job, you would be left up the creek without a paddle.

[c4757p]

It's too bad that the companies don't have mentor programs, but it is equally sad that the students don't have the motivation to pursue extracurricular experience before they graduate.

Sad that you blame the students while the real blame actually IMO lies with the colleges and universities that skimp on the education and society that pressures students to graduate asap.

This is why I said extracurricular. So far, with the exception of Semiconductor Physics, I have yet to take a class which I did not already know in depth before the class started, because I'm studying something that interests me and I put in the effort to teach it to myself before I even study it in school.

When I was studying CS, a good two thirds of the students were like that. There were many somewhat proficient programmers in my freshman Intro to CS class, and they improved as the curriculum became more advanced, always ahead of the curve. In EE, I am the only one I have met who does that.

And if you want to "blame" society, that's fine, then society is the reason they have no motivation. But the point stands that they do not, and that I think this is sad. Not sad in a sarcastic way, just frowny-face sad.

In my electronics lab, I work with a kid who can't look at a pinout diagram and figure out what goes where , and he's quite content to sit back and just let me do it. Part of me doesn't want to be his lab lackey, but the other part understands that if I try to force him to do more than punch buttons on the scope, we'll be there all damn day . At least he seems to be genuinely interested in some parts of the field, so I talk about them as much as I can and try to teach him practical knowledge about the circuits, and hopefully I can instill some desire to learn even more. Seems to be working, a bit, I think

Apologies for the thread derailment, I'll shut up now. Back to capacitors?

[george graves]

Yes this is quite a problem if you are not aware of this, that is why I suggested it as a subject for a fundamental friday:

Good idea.  Hope Dave does it.

[Kjelt]

Ok, I'm curious.  Did anyone with a BSEE or MSEE actually have a college class (or lab) that talked about the different characteristics of real-world capacitor types? 

Come on you must have had the difference in ESR, frequency response between elco's, tantals and ceramics? Maybe the differences between MKT, MKP and the likes?
We didn't have much SMT unfortunately, was not a main topic back then.
Maybe it also depends on the graduation topic. In our schoolsystem we have a couple of years general classes, lots of math and theory in them, then you choose the direction you like to graduate and get more in depth knowledge of that topic and surroundings. Unfortunately I choose Technical Computer Science so I had a lot of digital electronics theory but too little of the analog stuff

And if you want to "blame" society, that's fine, then society is the reason they have no motivation.

Hold on I didn't say that. I wonder how a student without motivation can still graduate, explain that to me? Because if that is the case then there is something wrong with the educational system don't you think? There could be tons of reasons for this. From my own experience I remember some classes where the first lecture was given by a professor that came from the industry, he held a very good lecture that made everyone enthousiastic and energetic to follow and pass the course. The rest of the classes were given by an assistant that just read page and page of the textbook and all the energy was gone.
Another class was given by an academic professor that the entire hour just wrote formula's on the board and showed how a great mind he was. No energy there.
So what I am trying to say is that if there is more then 10% unmotivated students, how can you blame the students? Why are they unmotivated and why is the university/college not stopping them after a semester if they have no motivation?

But the point stands that they do not, and that I think this is sad. Not sad in a sarcastic way, just frowny-face sad.

I agree, very sad.

In my electronics lab, I work with a kid who can't look at a pinout diagram and figure out what goes where , and he's quite content to sit back and just let me do it.

I understand that this is the easiest way out for you and I do not want to judge, but do you see that in this way you are part of the system? I don't know the person so maybe if you enthousiastic try to transfer knowledge he  panics , on the other hand maybe you give him that sparc of energy that he needed to go read more about his job en study and go for example to this forum and show him there is more then just turning knobs and magical things happen 

At least he seems to be genuinely interested in some parts of the field, so I talk about them as much as I can and try to teach him practical knowledge about the circuits, and hopefully I can instill some desire to learn even more. Seems to be working, a bit, I think

Great , that is what I mean, I'll take my hat of for you sir. 

Apologies for the thread derailment, I'll shut up now. Back to capacitors?

  yep same here, let's get back on topic.

[c4757p]

Not to continue this more, but since you asked me a few questions, I'll slip in answers... I really did intend to comment on the topic.

Of course any student who graduates has at least the motivation to do exactly as he is told. That's not useful motivation, he might as well do that for the manager at the local McDonalds. I meant internal motivation to learn about things that interest him.

Again, I never said I "blame" them for being unmotivated. If there are more than 10% unmotivated students, there still are unmotivated students, which is all I said. And I don't think the university is to blame - the students are adults for Christ's sake. We ought to be able to expect that once a child reaches adulthood, he'll have a somewhat healthy mindset about learning and work ethic. A motivated character has to be shaped all throughout childhood, and if the blame is to rest anywhere, it is with the society that allows children to be pushed through grade school the way they are, as little test-taking machines: supposed to be 250% "motivated", but only to do exactly as they're told, screw any personal curiosity and passion (and for the "gifted" ones, milk as much extreme effort out as you can at a young age so they're burned out by 18).


Now, to contribute to the thread, because this is interesting to me as well:

While anyone who really depends on capacitors regularly will eventually develop an intuition for just how much capacitance is left at a given bias, I really, really wish manufacturers would start including this data in the datasheet!! It's not like it's hard to measure, I've done it myself on an LCR bridge with a significantly larger bias-blocking cap and bias fed in through a 1meg resistor... But even though that Maxim document says to always read the datasheet, lots of datasheets don't even specify voltage coefficient!

[jmaja]

While anyone who really depends on capacitors regularly will eventually develop an intuition for just how much capacitance is left at a given bias, I really, really wish manufacturers would start including this data in the datasheet!! It's not like it's hard to measure, I've done it myself on an LCR bridge with a significantly larger bias-blocking cap and bias fed in through a 1meg resistor... But even though that Maxim document says to always read the datasheet, lots of datasheets don't even specify voltage coefficient!

That is so true! I first run into this problem last spring (no, I haven't studied EE, just Physics including one course in electronics, but quite a few courses on quantum physics including what happens in e.g. a PN junctions at that level): https://www.eevblog.com/forum/projects/ceramic-capacitor-dimension-for-a-buck-converter/msg224126/#msg224126

I had read several cap datasheets very carefully and yet this was a total surprise for me. And it wasn't at all easy to find the manufacturer provided more detailed data. There were no links to them from their websites etc. I didn't get any feedback when I asked about it from online questionary. Once I understood to put the keyword "spice" in search engines I started to find out the online calculators and program downloads.

Still I don't know where is this data for e.g. Samsung capacitors, which I would otherwise have used in some cases.

Another thing about ceramic caps: Is there a way to know beforehand will there be audible noise or not? Can you choose a ceramic one (still with high capacitance and low ESR), which wouldn't make noise. One of my buck converters makes a bit annoying noise, but only at very low load. There was a warning about this in the converter datasheet, thus it didn't come as a surprise.

[c4757p]

Another thing about ceramic caps: Is there a way to know beforehand will there be audible noise or not? Can you choose a ceramic one (still with high capacitance and low ESR), which wouldn't make noise. One of my buck converters makes a bit annoying noise, but only at very low load. There was a warning about this in the converter datasheet, thus it didn't come as a surprise.

I'd just get a bunch of samples and test them! Drive them at the highest ripple current you can do without damaging them through a range of expected operating frequencies and see if they do anything.

Are you sure it's not the inductor?

[jmaja]

Are you sure it's not the inductor?

No. They are side by side, so it's hard to tell. But seems most likely, since the datasheet says: "The LTC3630’s burst frequency depends on the load current, and in some applications at light load the LTC3630 can excite the ceramic capacitor at audio frequencies, generating audible noise. If the noise is unacceptable, use a high performance tantalum or electrolytic capacitor at the output."

[Zero999]

I've heard about this before.

Here's an article I read awhile ago about using a ceramic capacitor as a varactor. It's probably not much use as a practical VCO, unless it's used to make something like a siren which isn't critical.
http://www.edn.com/design/analog/4334383/Multilayer-capacitor-doubles-as-varactor

[westfw]

Come on you must have had the difference in ESR, frequency response between elco's, tantals and ceramics? Maybe the differences between MKT, MKP and the likes?

Nope; never.  I mean lossy dielectrics were part of the theoretical physics in the EM class(es), but I don't recall it ever being related to practical circuit design. (OTOH, that was a long time ago, so my memory could be faulty.)

It COULD have been a good topic for the lab classes, but those were rather useless (IMO) collections of "build this circuit using the components provided and learn to use various pieces of lab equipment by measuring the behavior and writing it up."  And the usual measurements were of semiconductor characteristics.

Has anyone seen an online class covering real-world component selection?
I took a look through MITs open courseware and didn't see anything that jumped out, for example.

(I think you can blame a lot of the current neglect on the prevalence of digital circuits.  People don't actually need a 10uF 6V cap, they need "a cap that is small enough to sprinkle around the power pins of my microprocessor, and will adequately bypass noise on the power rails."  As the requirements get tighter, the microprocessor datasheet gets more specific about "use THIS capacitor and put it HERE", and the actual operating point values are a bit irrelevant.  If you DO have an analog segment of your design, you're still following an app note that warns you when cap parameters are particularly important, or tells you what type of cap to use.)

[c4757p]

(I think you can blame a lot of the current neglect on the prevalence of digital circuits.  People don't actually need a 10uF 6V cap, they need "a cap that is small enough to sprinkle around the power pins of my microprocessor, and will adequately bypass noise on the power rails."  As the requirements get tighter, the microprocessor datasheet gets more specific about "use THIS capacitor and put it HERE", and the actual operating point values are a bit irrelevant.  If you DO have an analog segment of your design, you're still following an app note that warns you when cap parameters are particularly important, or tells you what type of cap to use.)

So in other words, they're training a generation of useless engineers who are only capable of crapping out the most popular designs, and must defer to a real engineer who wrote them a nice, shiny app note when they have to do something out of their comfort zone.

Sounds about right.

[Neilm]

When I went to uni, the course was mostly about digital systems. Analogue was covered, but it only really went for op-amp circuits. Mostly assuming the signal was going into an ADC. Lots of control theory, maths, Fourier and Z transforms, IC design but I don't recall anything about component constructions. I don't even think there was anything on different caps or resistor types.

Result being I learned the hard way that caps have voltage co-efficients, vary with temperature and the like.