EMRFD Message Archive 2907
Message Date From Subject 2907 2009-04-10 11:49:01 Stephen Wandling Choosing capacitors Paul, NA5N posted this on the QRP-L list some time back.
-------- Original Message --------
Subject: Re: [QRP-L] choosing capacitors
Date: Sun, 01 Mar 2009 23:40:39 -0700
To: Kendrick KB1NCR <email@example.com>
CC: QRP-L list <firstname.lastname@example.org>
Kendrick KB1NCR writes:
> I have NO IDEA why I would choose one kind over another for a QRP application.Kendrick and others,
> Is there a rule of thumb or any hard science, any text book chapter I
> can read to get some practical incite?
Funny this capacitor thread popped up. The ARRL has asked me to give them a
hand reworking the Components chapter in the next Handbook. I have been
working on the capacitor section for two days, and most of today, trying to
figure out how to make capacitor selection easier to understand. That is,
how to answer exactly the questions (and confusion) you posed. You're not
Capacitors clearly comprise the largest choices of component selection. All
of the varieties and specs can seem to be daunting. There's NPO, X7R, Y5V,
5%, 10%, ceramic disc, multilayer ceramics, multilayer films, metalized
films, silver micas, disc caps, encapsulated, epoxy coated ... well, you get
the drift (pun intended). The EIA (Electronics Industry Alliance) has
actually done a good job making capacitor selection easy, but seemingly
overlooked in most literature.
Look at your DigiKey or Mouser catalog. For most capacitor types, whether
ceramics, films, poly-something or another, the catalogs will list either
the tolerance (5%, 10%, 20%) or Class (Class 1, 2, 3 or 4), or the
temperature codes (NP0, X7R or X5U). Understanding these three things
answers 95% of your questions.
CLASS 1 are at least 5% tolerance capacitors. These are the most stable in
tolerance, long term drift and temperature coefficient. NP0 (same as the
EIA designation C0G) is the dominant example of a Class 1 capacitor. Class
1 caps should be used in sensitive frequency determining circuits, such as a
VFO or tuned circuit for stability and minimal drift over temperature. It
makes very little difference whether a particular Class 1 5% capacitor is
ceramic or a film type. Only the exotic polypropelene 2% tolerance caps
(and other expensive types) are a bit better, but all still Class 1 for the
top level of stability.
CLASS 2 are 10% tolerance caps with a fairly stable temperature coefficient
(but not as good as Class 1). A cap with a temperature code of X7R is an
example of a Class 2 cap. They can be used in frequency determining
circuits where it's drift vs. temperature is not overly critical ... such as
the feedback and tuning caps in a crystal oscillator or in many tuned
circuits. Most tantalum caps are also 10% tolerance Class 2.
CLASS 3 are also 10% tolerance caps, but with a worse temperature
coefficient than Class 1 or Class 2. Y5V and Y5U are two common Class 3
capacitors. They should not be used in sensitive frequency determining
circuits. They are generally cheaper than Class 1 and very suitable for
bypass caps, interstage coupling, and most other uses in a circuit.
CLASS 4 are the least preferred with tolerances of 20% or more. With modern
manufacturing techniques, it is rare to find a Class 4 capacitor today. The
exception are the aluminum electrolytic caps. These are generally Class 4
20% tolerance, which is more than suitable for their intended applications.
(You're not making tuned circuits with electrolytics, are you?)
So, to make it even simpler, here's what to remember or jot down:
Class 1 = 5% or better = NP0/C0G = the best class of caps
Class 2 =10% = X7R = next best class of caps
Class 3 =10% = Y5U/Y5V = common, run-of-the-mill caps
Class 4 =20% = don't use unless an electrolytic
NP0 is actually an obsolete designation. The old temperature codes, such as
P50 for +50ppm, N100 for -100ppm, or NP0 for zero ppm/degree C, are no
longer used, replaced by the EIA codes, C0G, X7R, etc. However, the NP0
designation is so engrained as the "zero drift capacitor," most
manufacturers continue to use the designation. QRPers certainly still do!
IF YOU'RE BUYING NEW ...
Select Class 1 (NP0/C0G) for caps in frequency determining circuits.
Select the cheaper Class 3 for most all other applications.
IF YOU'RE USING OLD JUNK BOX/SURPLUS CAPS ...
For frequency determining caps, use only if you know it is an NPO type. NP0
ceramic discs usually have a black stripe or paint blob at the top.
Assume all other capacitors in your junk box are Class 3. Use them for
bypass caps, interstage coupling, etc., but not as the frequency determining
caps in a VFO. If you happen to have some really old paper and wax tubular
caps, or those really nifty JAN postage stamp caps, you can figure they're
at least 20% off the marked value by now, and thus a Class 4. They might be
great for impressing the FDIM buildathon judges, however.
So get out your DigiKey or Mouser catalog and put this to use. Most pages
will indicate whether the cap is an NP0, X7R or Y5U/Y5V (most common types).
If it doesn't, look at the tolerance column. If it's 5%, it's a Class 1; if
it's 10%, assume it is Class 3 unless otherwise stated. Of course, many
pages also say "Class 1 or Class 3." The tolerance column, 5% or 10%, will
tell you whether it's Class 1 or Class 3.
Again, use Class 1 for frequency determining caps; Class 2 or 3 for most all
So what about the ceramic, metalized film thingie? It really doesn't make
much difference. A Class 1 is a Class 1, regardless of the dielectric used.
Ceramics are the most stable, but the manufacturing process makes them a bit
more expensive. All the multilayer film types are just different methods to
manufacture a cap that has a certain capacitance and generally in a smaller
package and cheaper compared to ceramics. Most surface mount caps are
multilayer films with a ceramic coating. This smaller film size vs.
ceramics also makes some of them sensitive to temperature changes. This is
why you check to see if the film caps are a Class 1, 2 or 3 as the
application requires. And finally, some multilayer film caps use special
coatings to achieve very accurate capacitances and improved temperature
coefficients. These will be Class 1 and can be significantly more expensive
than ceramics (like polycarbonates or silver micas). But, they're well
worth the price if you're putting something into orbit :-)
The one exception I can think of to the above is to use a Class 1 or 2 cap
with a low ESR in active filter audio circuits. You want a cap whose value
will change very little over time (Class 1 or 2) and has a low series
resistance (ESR). A high ESR cap in an active filter (which usually have
unity gain) will gobble up much of the audio signal rendering the circuit
nearly worthless. The difference between good caps and mediocre caps in an
active filter is like night and day.
Hopefully, I will have some useful illustrations and tables to show all of
this for the handbook. The ARRL does seem to be motivated in changing the
Components Chapter in the handbook to be more useful for homebrewers, and to
encourage new homebrewing. If you have any ideas on what should be changed,
added, or presented in a different way, let me know. I have already
submitted a fair number of suggestions ... like replace the 3 pages of light
bulbs with some LEDs and good toroid information. So far, I have received
good support from Newington to get the Components chapter back to a good and
updated reference source. I know other chapters are getting the same level
of attention (with several notable QRPers involved!).
My next head scratcher will be how to make the toroid and semiconductor data
more concise and useful.
Anyway, Kendrick, I hope the above answers most of your questions. Consider
your junk box caps as 10% Class 3 general purpose caps unless you know
72, Paul NA5N (Class 3)
QRP-L mailing list
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Hunger is not caused by a scarcity of food
but a scarcity of democracy - Frances Moore Lappé
2908 2009-04-10 15:07:13 Russell Shaw Re: Choosing capacitors Stephen Wandling wrote:
> Paul, NA5N posted this on the QRP-L list some time back....
> The one exception I can think of to the above is to use a Class 1 or 2 capIt might be a good idea to rate types in terms of Q factor, because being
> with a low ESR in active filter audio circuits. You want a cap whose value
> will change very little over time (Class 1 or 2) and has a low series
> resistance (ESR). A high ESR cap in an active filter (which usually have
> unity gain) will gobble up much of the audio signal rendering the circuit
> nearly worthless. The difference between good caps and mediocre caps in an
> active filter is like night and day.
> Hopefully, I will have some useful illustrations and tables to show all of
> this for the handbook. The ARRL does seem to be motivated in changing the
> Components Chapter in the handbook to be more useful for homebrewers, and to
> encourage new homebrewing. If you have any ideas on what should be changed,
> added, or presented in a different way, let me know. I have already
> submitted a fair number of suggestions ... like replace the 3 pages of light
> bulbs with some LEDs and good toroid information. So far, I have received
> good support from Newington to get the Components chapter back to a good and
> updated reference source. I know other chapters are getting the same level
> of attention (with several notable QRPers involved!).
> My next head scratcher will be how to make the toroid and semiconductor data
> more concise and useful.
> Anyway, Kendrick, I hope the above answers most of your questions. Consider
> your junk box caps as 10% Class 3 general purpose caps unless you know
inversely proportional to esr, the numbers are more dramatic, and useful for
It would be a good idea to single out the highest Q caps. The highest Q by
far that i've found apart from an air capacitor, are those old clear cylindrical
polystyrene caps with the really thin and long leads.
Another important effect of esr is that it sets the power handling capacity (no
pun!) for transmitter circuits, or in SMPS circuits using electrolytics.
The self-inductance is important for obvious reasons too.
2909 2009-04-10 15:52:21 Chris Trask Re: Choosing capacitors >for
> It might be a good idea to rate types in terms of Q factor, because being
> inversely proportional to esr, the numbers are more dramatic, and useful
> tuned circuits.cylindrical
> It would be a good idea to single out the highest Q caps. The highest Q by
> far that i've found apart from an air capacitor, are those old clear
> polystyrene caps with the really thin and long leads.Amen. They were made by Mallory and others, and they are/were the best
money can/could buy. My first choice these days are the BF series
metallized polyester types that were earlier made by Thomson but which have
been superceded by the lead-free BQ series made by AVX (available through
,----------------------. High Performance Mixers and
/ What's all this \ Amplifiers for RF Communications
/ extinct stuff, anyhow? /
\ _______,--------------' Chris Trask / N7ZWY