John Lawvere
I'm skeptical of black-box capacitance meters. Connect a square-wave source, resistor, and capacitor in series, with grounded end of source? and one end of cap in common, and watch charging/discharging of cap voltage?with an oscilloscope. Measure half-life of exponential decay, calculate time constant, and solve?for capacitance. Remember that total resistance is sum of resistor you put in circuit plus generator resistance of square wave source. If you plan to do this a lot--especially with small caps--you should do some "calibration" runs with several accurately known small caps: find capacitance of o'scope probe. On Monday, August 3, 2015 9:39 AM, "Jerry Lee Marcel jerryleemarcel@... [LTspice]" wrote:
?
Le 03/08/2015 17:58, zio_bapu@...
[LTspice] a ¨¦crit?:
?This has already been answered. When the real capacitor is close enough to the basic "one C- two R" model, they generally can be trusted, although different measurement frequencies are likely to give slightly different values. As for electrolytics, they cannot be trusted to give an accurate value, because such a concept does not exist; as I mentioned earlier, their capacitance varies with frequency, but also electrolytic show significant DA (dielectric absorption), a non linear phenomenon that makes the apparent capacitance change over time. In addition, capacitance decreases with time, but some regeneration can happen, and it also varies significantly with temperature. These issues are known from the beginning, that's why electrolytics are not to be used in applications where their value is critical, as in filters (although many cheap loudspeaker manufactures use NP lytics in their passive x-overs). From their inception, designers have commonly used a factor 10.
|