29-03-2019, 07:54 AM
Without doing the sums I suspect that you can equate ESR to leakage at a particular frequency. Though this must break down completely as you approach zero frequency. It's also not terribly helpful. Any non-electrolytic capacitor you're likely to use in most applications will have almost infinite leakage resistance, something that won't matter in most applications but may be important in (for example) a long period sample and hold. ESR may matter in some applications, such as pulsed power, but hardly matters for audio coupling.
For electrolytics you always have to tolerate some leakage but it hardly matters on a smoothing cap over a very wide range. ESR does matter in many cases and will make the difference between a circuit working or not.
So while it may be possible to equate these 2 parasitic resistances at a given frequency I can't see that it's much use to anyone. If you measure the ESR of a reservoir cap it may be important. While leakage won't be, within reasonable limits.
For electrolytics you always have to tolerate some leakage but it hardly matters on a smoothing cap over a very wide range. ESR does matter in many cases and will make the difference between a circuit working or not.
So while it may be possible to equate these 2 parasitic resistances at a given frequency I can't see that it's much use to anyone. If you measure the ESR of a reservoir cap it may be important. While leakage won't be, within reasonable limits.
www.borinsky.co.uk Jeffrey Borinsky www.becg.tv








