I read somewhere (a electronics/music forum) that the higher rated voltage capacitors may have closer tolerances........but there was nothing to support this, so I thought id as here.
not true
The material the capacitor is made of has the biggest influence on the tolerance. How it's made is the next biggest influence.
It's pretty hard to make a really big, really accurate capacitor, so you don't find a 1.0000000 farad capacitor.
Most high accuracy capacitors are made for low voltage circuits, but at least in theory, you could make a high voltage high accuracy capacitor.
But a .01ufd at 50V isn't going to be inherently more accurate than a .01ufd 20V capacitor.
a .01ufd at 50V isn't going to be inherently more accurate than a .01ufd 20V capacitor.
Thank you brtech, that was my highly uneducated feelings on the matter.
you don't find a 1.0000000 farad capacitor.
Plenty of 1 farad caps used for power supply smoothing in aftermarket car-audio installs, but I wouldn't trust them to be of high quality!
Plenty of 1 farad caps used for power supply smoothing in aftermarket car-audio installs, but I wouldn't trust them to be of high quality!
True, but they won't be exactly 1.0000000 farads.
There are a couple of bits of truth here, sort of...
-
The very-low voltage capacitors (say, 10V and less) are generally trying to squeeze the maximum capacitance into the minimum space, and may be sacrificing close tolerances to do so. It wasn't uncommon for Aluminum electrolytic capacitors in general to have tolerances of -10/+80%, for example, which was OK because in the typical applications it was "more uF is better."
-
The latest generation of exotic ceramic dielectrics (X5R, Y5V?), which permit things like 100uF 6V caps in an 0805 footprint (!) apparently have "quirks" that affect the tolerance (things like capacitance being unusually dependent on applied voltage, for example.)
So while it's not true that voltage rating has a direct effect on accuracy, it MAY be true that a capacitor whose voltage rating is particularly close to the actual applied voltage has a less accurate value than a more conservatively specified capacitor...
Read your data sheets! If you can find them (some of the linked DS's at digikey didn't have much more than physical measurements 
)
That is true for ceramic capacitors like X7R, X5R and Y5V.
There is a loss of capacitance as the voltage increases. For the
X7R and X5R the tolerance is around 30% at the rated voltage.
For Y5V the tolerance is around 80%.
Using an X7R at 40% or 50% of its rating gives you essentially
a flat capacitance (which is a tighter tolerance). For ceramic
caps I try to stick to X5R and X7R (for caps 1000pF and up).
If I use a Y5V are I will typically double or triple the voltage rating.
For ceramic capacitors take a look at the dielectric datasheets at
Kemet.
(* jcl *)
www: http://www.wiblocks.com
twitter: http://twitter.com/wiblocks
blog: http://luciani.org
A couple of poster have implied (but not stated) a good point: if one device is listed as 1.00uF and the other is listed as 1uF, it is implied that the 1.00uF is a tighter tolerance than the 1uF (but always check the datasheet to see if the manufacturer knows the rules about significant digits).
-j
Or it could just a a quirk of their database integer format... 
Most passive components like capacitors actually state their tolerance deviation and as stated above already, it's not a factor of voltage. While it is true that you might get longer "life" out of a higher voltage capacitor... since you aren't stressing it... it will still behave the same as a similar lower voltage unit. That is, unless you were stressing the lower voltage device beyond stated limits.
Sometimes though... tolerance deviations of up to 10% or more may not impact your circuit very much... such as a power supply filter. In tuned circuits like those in used in a radio perhaps... the value of a capacitor might need to be pretty close to the exact value printed on it and you would pay more to get the 1% to 5% tolerance version.
This thread begs the question of why you want high tolerance capacitors anyway. About the only reason would be for the design of multi pole filters.