Go Down

Topic: smd capacitors (Read 735 times) previous topic - next topic

tinman13kup

I was looking up the finer differences between the X7R and X5R capacitors when I came across this voltage deviation . I must admit, I was a bit surprised by what I read, and if this article is true, the implications on the hobbyist putting together boards.

  I can fully understand a cheap capacitor falling into a +/- 20% tolerance, but it appears the issue is outside the scope of that tolerance.

  When I order a 10uF 10% 16V capacitor, I'm not expecting to get 4uF performance.

  Has anyone had circuit performance degraded because of this? Is there any manufacture who seems to be better at maintaining the advertised capacitance when it's within its operating specs?

  I understand the article leads to saying the larger case sizes seem to be closer to capacitance, but I don't want to be stuck using >1206+ to achieve specific capacitances for some of the chips.
Tom
It's not a hobby if you're not having fun doing it. Step back and breathe

MarkT

High value ceramic capacitors use ferro-electric materials like BaTiO4, and are very non-linear,
this should be a well known thing - its one of the reasons you don't use them in analog filters
for instance, but use plastic film instead.  For decoupling its fine, since the capacitance value is
not at all critical so long as its big enough.

There is nothing anything a manufacturer can do to make ferroelectric materials behave
linearly, any more than with ferromagnetic materials, its a fundamental limitation of the material,
but there are different formulations with different amounts of non-linearity - but there are
tradeoffs...

The rated capacitance is the small signal value, the rated voltage is the maximum that won't
cause damage, the non-linearity starts well before that.

Note that not all ceramic caps use ferroelectric materials, its only ferroelectrics that are
grossly non-linear like this, but ferroelectric materials are used when high values of capacitance
are wanted because they have dielectric constants in the thousands, not just 5 or 10.

https://en.wikipedia.org/wiki/Ceramic_capacitor#Class_2_ceramic_capacitors
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

larryd

@MarkT
How do you link to that paragraph on the web site.
i.e. to:       
Class 2 ceramic capacitors

.
No technical PMs.
If you are asked a question, please respond with an answer.
If you are asked for more information, please supply it.
If you need clarification, ask for help.

tinman13kup

That does look handy (linking to paragraph)

Maybe I'm reading too far into this. I was understanding this; that the X7R rating doesn't take the voltage/capacitance into account, only that whatever it might be only changes < +/-15% for temp variation. The non-linearity of the capacitance has no standard, only that it "is", and it's up to the end user to dig past the datasheet to find out what that non-linearity vs V might be if they want to get a capacitance even close to what they were looking to get at a specified voltage.

  The graphs he posted with the various size, voltage rating, and type -vs- applied voltage doesn't state any temp variation, only the non linearity of applied voltages. It's a little hard to pick various ones out of the legend, but it looks to me that all those tested only keep their stated capacitance for 10% of their rated voltage, then start deviating rapidly, with some faster than others.

  In general, yes, I use them for decoupling, but is using the 10% a good practice to use all around?
Tom
It's not a hobby if you're not having fun doing it. Step back and breathe

INTP

@MarkT
How do you link to that paragraph on the web site.
i.e. to:       
Class 2 ceramic capacitors

.
On Wikipedia, clicking on the heading within the table of contents at the top of the page to jump down to that section will give you the url for the jump

larryd

#5
Nov 27, 2016, 03:58 am Last Edit: Nov 27, 2016, 04:03 am by LarryD
Thanks INTP.

I was hoping there was a search function so one would be able to link to a heading such as 'Example output for Uno' on this page.
http://www.gammon.com.au/forum/?id=11635


No technical PMs.
If you are asked a question, please respond with an answer.
If you are asked for more information, please supply it.
If you need clarification, ask for help.

allanhurst

All the high dielectric ceramics are awful, both in temperature characteristics and high frequency loss...

OK for decoupling.

C0G  are good but you can't get high values...

regards

Allan

tinman13kup

The training I received was for all through hole components and tubes, so I was instinctively correlating smd's to military grade through holes. I didn't get anything in digital, so reading through more complex datasheets takes a bit more time for me to fully comprehend. Some of the equations used make me scratch my head, while others ring a bell from long ago.

 Now is the time for me to learn habits with smd's, whether good or bad. I'd like to avoid disappointments by sticking to good ones.
Tom
It's not a hobby if you're not having fun doing it. Step back and breathe

MarkT

@MarkT
How do you link to that paragraph on the web site.
i.e. to:        
Class 2 ceramic capacitors

.
By referencing the relevant fragment?  

scheme:[//[user:password@]host[:port]][/]path[?query][#fragment]

https://en.wikipedia.org/wiki/Uniform_Resource_Locator#Syntax
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

larryd

Thank you MarkT, very much appreciated.


.
No technical PMs.
If you are asked a question, please respond with an answer.
If you are asked for more information, please supply it.
If you need clarification, ask for help.

Go Up