CR4® - The Engineer's Place for News and Discussion®


Previous in Forum: DVR Password Recovery   Next in Forum: PCM for Battery Pack
Close
Close
Close
14 comments
Guru

Join Date: Aug 2005
Location: Hemel Hempstead, UK
Posts: 4140
Good Answers: 230

Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/27/2016 11:51 AM

This is part of a datasheet for a well known make of MLC

The impedance and ESR curves are pretty much what I'd expect, but the capacitance versus frequency curve came as a bit of a shock.

What does it really mean?

I can't find a similar graph on any other manufacturers websites.

__________________
We are alone in the universe, or, we are not. Either way it's incredible... Adapted from R. Buckminster Fuller/Arthur C. Clarke
Register to Reply
Pathfinder Tags: capacitance graph MLC
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Comments rated to be Good Answers:

These comments received enough positive ratings to make them "good answers".

Comments rated to be "almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, rate them!
2
Guru
Engineering Fields - Electrical Engineering - Been there, done that, still doing it. Engineering Fields - Control Engineering - New Member

Join Date: Dec 2008
Location: Long Island NY
Posts: 12219
Good Answers: 774
#1

Re: Capacitance versus frequency for multilayer ceramic capacitor.

04/27/2016 12:16 PM

This is the effect found when one exceeds the resonant frequency of a real capacitor. The parasitic inductors end up dominating the impedance. The second graph is showing the effective capacitance of that 20 uF component as it approaches, reaches and then slightly exceeds the resonant frequency.

__________________
"Don't disturb my circles." translation of Archimedes last words
Register to Reply Good Answer (Score 2)
3
Guru

Join Date: Apr 2010
Posts: 5752
Good Answers: 576
#2

Re: Capacitance versus frequency for multilayer ceramic capacitor.

04/27/2016 12:23 PM

A physical capacitor is actually a capacitance in series with the inductance of the leads. In this case the inductive reactance is numerically equal to the capacitive reactance at 1 MHz. At higher frequencies, the inductive reactance predominates and the reactance increases with frequency rather than decreases.

Register to Reply Good Answer (Score 3)
Guru

Join Date: Mar 2007
Location: at the beach in Florida
Posts: 19263
Good Answers: 1129
#3

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/27/2016 2:31 PM

Looks like it hit a wall at just under 1 Mhz....

..."Most discrete capacitor types have greater or smaller capacitance changes with increasing frequencies. The dielectric strength of class 2 ceramic and plastic film diminishes with rising frequency. Therefore, their capacitance value decreases with increasing frequency. This phenomenon is related to the dielectric relaxation in which the time constant of the electrical dipoles is the reason for the frequency dependence of permittivity. "...

https://en.wikipedia.org/wiki/Ceramic_capacitor

__________________
Life is like riding a bicycle. To keep your balance you must keep moving. A.E.
Register to Reply
Guru

Join Date: Aug 2005
Location: Hemel Hempstead, UK
Posts: 4140
Good Answers: 230
#4

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/27/2016 3:00 PM

Thanks all three of you:-

Now I'm very perplexed, it looks as though all these capacitors:-

https://product.tdk.com/en/search/capacitor/ceramic/mlcc/list#1shape%5B%5D=1010101&1cap%5Bunit%5D=1.00E-06&1cap%5Bt%5D=22&1captol_sbl%5B%5D=M&1rv%5Bcomp%5D=%3E%3D&1rv%5Bdc%5D%5B%5D=4&_l=20&_p=1&_c=part_no-part_no&_d=0

Are useless as capacitors above about 1MHz.

Even the low ESL reverse geometry 0306 2µ2 ones give up at 10MHz.

I had always believed that MLCs were good for high frequency decoupling, and, that was why we scattered them all over our designs.

I guess I would be carrying on in blissful ignorance if I hadn't come across this graph, and, as I said no other manufactures seem to include it.

I'm obviously missing something here.

__________________
We are alone in the universe, or, we are not. Either way it's incredible... Adapted from R. Buckminster Fuller/Arthur C. Clarke
Register to Reply
Guru
Engineering Fields - Electrical Engineering - Been there, done that, still doing it. Engineering Fields - Control Engineering - New Member

Join Date: Dec 2008
Location: Long Island NY
Posts: 12219
Good Answers: 774
#5
In reply to #4

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/27/2016 3:36 PM

Without knowing the in depth details of your circuit, it would just be a WAG if the decoupling capacitors are properly selected or if they are just a minor waste of money that's not harming the circuit.

__________________
"Don't disturb my circles." translation of Archimedes last words
Register to Reply
Guru

Join Date: Aug 2005
Location: Hemel Hempstead, UK
Posts: 4140
Good Answers: 230
#9
In reply to #5

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/28/2016 6:10 AM

It's not the circuit I'm looking at, at the moment I'm perplexed by: it's every microprocessor or other high speed digital circuit I've seen in the last 40 years.
As I said I'm obviously missing something here.

__________________
We are alone in the universe, or, we are not. Either way it's incredible... Adapted from R. Buckminster Fuller/Arthur C. Clarke
Register to Reply
Guru

Join Date: Sep 2012
Location: Iowa, USA
Posts: 565
Good Answers: 48
#6
In reply to #4

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/27/2016 4:45 PM

I was shocked when I first started working on fixing radiated emissions from avionics that I was qualifying. And I learned about a vector impedance meter that the lab had (HP 4815A). We generally used NPO ceramic capacitors. So we made a chart of the various ones available from engineering stock (self resonance vs Farad). Found out real quick that 5000pf and smaller is all that would work. At that time the processor clocks were 10MHz, so 3rd and 5th harmonics is what we were trying to quiet.

The bigger lesson, is just spreading them (spray of decoupling capacitors) across the PC board does not do the best job of quieting the noisy parts. They make the PWB quiet in the power and ground plane locations where they are mounted, but since the planes were "swiss cheese" from all the clearance holes for the thru's, which effectively made the copper plane a good inductor, this did not properly bypass the digital microprocessor logic devices.

Think of a 8 or 16 bit buffer chip, and the simultaneous output totem pole drivers having to charge the bus (addr or data or micro if you don't buffer your bus). Same with memory chips driving the bus. How do you decouple them to source this instantaneous charging of the stray capacitance of the PWB traces? As well the input capacitance of the logic that is being driven.

We found that typically 3 capacitors (of 'large' to small capacitance) were needed per device that drove 8 bits or more at the processor clocking speeds. And how these were connected from the power/gnd planes to the device Vcc/gnd pins was critical to RF emissions compliance.

You need to figure out how to minimize the inductance of the circuit path, and make the decoupling capacitors source this instantaneous charging current. In our case we took circuit traces from the planes to the capacitors then to the device power pins. Making the ground trace as short as possible to minimize ground bounce. And the Vcc trace as wide as possible, as it will have some path length, so minimize it's inductance.

Most of the emissions were all conducted on the low volt power rails, then through the switching power supply where the aircraft power wire (external to the UUT Unit Under Test) became and antenna radiating off the RF noise.

__________________
ignator -
Register to Reply
Associate

Join Date: Aug 2015
Posts: 52
Good Answers: 2
#11
In reply to #6

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/28/2016 6:37 AM

I retired from the Army in 1983, and started working in electromagnetic compatibility engineering. Since I'd already been playing with electricity and electronics since age 12 (that was when I built a coil gun in my bedroom, shooting ball bearings into the wall) I was familiar with most of these effects already, the advantage of having learned for myself a great deal of the physics behind them. I doubt it would surprise you if I said I had spent 15 years of my military career working on avionics.

Now retired again, I still hope for the occasional contract; a great many engineers don't actually understand the extremely basic physics behind what we are trying to do. Fixing problems is fun; preventing them is better!

Modern chip capacitors can provide more capacitance in smaller packages than older ones, and are therefore usable from lower frequencies through higher ones. It's incumbent on a design engineer to select capacitors appropriate to the application; this may require knowing which pins or balls in the package source or sink current for which of the outputs - and how many. These capacitors must supply (or sink) locally all the current needed to support the voltage on switching gates or devices during the transition time circuit board inductance retards the flow of current, otherwise, power and ground voltage fluctuations will show up on all the outputs has on-board EMI.

Even so, I never saw this level of detail provided in specification sheets.

Register to Reply
Guru

Join Date: Sep 2012
Location: Iowa, USA
Posts: 565
Good Answers: 48
#12
In reply to #11

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/28/2016 9:52 AM

I was a late bloomer with electronics. Digital was where I started. What's strange is most of the engineers hated running their product through qual test. I got to enjoy doing it. Mainly fixing others design issues. The product I worked on was initially for commercial avionics, but the Gov sys division, started putting it in military aircraft (non combat, mainly Fat-cat haulers). The product had 19 microprocessors, and keeping 'bird chirps' from being heard in VHF and HF bands was always an issue. The product was modular, where line replaceable modules were removed from a cabinet (card cage). They used 6 pins for ground plane reference. This made the ground return very inductive. So figuring out how to decouple the noise directly drove me to changing the way decoupling was implemented. Memory was MIL461 Army Part II emissions test, and RTCA DO160 Cat H emissions.

That and HIRF (high intensity radio freq.), and lightning compliance was my main design goal. I designed the cabinet and the protections for the modules which came from flight control, flight management, maintenance computer, and data switch groups. Then qualified the integrated system. It was "pulling teeth" to get any of the cognizant module engineers to show up and figure out how to fix their 'box'. Mostly had to do if for them.

__________________
ignator -
Register to Reply
Associate

Join Date: Aug 2015
Posts: 52
Good Answers: 2
#10
In reply to #4

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/28/2016 6:17 AM

No, they're not useless; they still block DC, and have relatively low impedance for some higher frequencies yet. The actual effect is that the current through the component no longer leads the voltage across it once the impedance becomes inductive - this is a function of the various package, via, and trace inductances of the board.

Register to Reply
Guru
Hobbies - DIY Welding - Wannabeabettawelda

Join Date: May 2007
Location: Annapolis, Maryland
Posts: 4179
Good Answers: 235
#7

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/27/2016 5:17 PM

Now, if you really want to get famboozled, look at capacitance vs. VOLTAGE curves. You may be shocked to find out that for some capacitors, the capacitance value drops way off as a function of voltage. The Y5V have this kind of characteristic

Also, one manufacturer's X7R or C0G or whatever may not behave the same as another manufacturer's X7R or C0G. You have to read the datasheets for the exact capacitor you choose. Don't let your purchasing change to an 'equivalent' without your approval, or you could get 'burned'.

Register to Reply Score 1 for Good Answer
Guru

Join Date: Apr 2010
Posts: 5752
Good Answers: 576
#13
In reply to #7

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/28/2016 11:59 AM

They vary with temperature also. And the same capacitor made by different manufacturers may behave differently.

https://www.maximintegrated.com/en/app-notes/index.mvp/id/5527

Register to Reply
Commentator

Join Date: Dec 2009
Posts: 83
Good Answers: 5
#8

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/28/2016 2:40 AM

That tropic was a great rememberance to the old days. In 1972 as a student I got my first experience with ceramic caps and Tantal elcos in transistor test circuits running at 4GHz! My mentor always told me: Due to self resonance all caps are scrap and only useable at lower frequencies. So we always had a bypass network with of special selected 100n, 10n ceramic caps followed by chip caps 470pf 47pF (not SMDs as these were not yet available but ceramic discs which were cut to order in our factory - but needed a slot in the PCB) This was followed by the bias network in front of the critical high frequency stages on the Teflon PCB consisting of a 5pf stub followed by a small (extreme expensive) ferrite choke and two 0.5pF stubs followed by with a PCB trace acting as RF Choke. It was always a big hassle to get the supplies really clean.

So it is interesting that this topic is still an issue half a century later!

Some laws are difficult to change!

Register to Reply
Guru

Join Date: Jan 2008
Location: Tucson, AZ
Posts: 858
Good Answers: 22
#14

Re: Capacitance Versus Frequency For Multilayer Ceramic Capacitor

04/28/2016 7:09 PM

SPECIFICATIONS are universal, it's product compliance to them that's NOT universal!

__________________
...and the Devil said: "...yes, but it's a DRY heat..!"
Register to Reply
Register to Reply 14 comments
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Comments rated to be Good Answers:

These comments received enough positive ratings to make them "good answers".

Comments rated to be "almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, rate them!
Copy to Clipboard

Users who posted comments:

70AARCuda (1); Brave Sir Robin (1); ignator (2); ka5s (2); Randall (2); redfred (2); Rixter (2); SolarEagle (1); uli_newBuilder (1)

Previous in Forum: DVR Password Recovery   Next in Forum: PCM for Battery Pack

Advertisement