TABLE OF CONTENTS Replacing the electronic charge with its numerical value and translating to more useful engineering units we can write for the shot current noise density:
As the shot noise dictates the ultimate precision with which the photocurrent can be measured, it is important to have a feel for its magnitude. Table 3.1 shows a few values.
| Current ( I p) | Shot noise current spectral density ( i n) | Relative noise currents i n: I p in 1 Hz BW |
|---|---|---|
| 1mA | | 1:55,000,000 |
| 1 A | | 1:1,754,000 |
| 1 nA | | 1:55,000 |
| 1 pA | | 1:1,754 |
This shows the precision that is possible in principle if the measurement system is shot noise limited. If we have 1mA of photocurrent, it should be possible to determine its magnitude with a precision of 1 in 55 million in 1Hz bandwidth, far beyond the resolution of the best analog-to-digital converters. In practice, many effects conspire to limit our precision to a value much worse that that.
Another big issue is what currents actually show full shot noise? This is a matter of some uncertainty in the literature. Netzer (1981) suggests that shot noise is seen in situations where charge carriers cross a barrier independently of one another, such as pn-junction diodes where the passage occurs by diffusion, a vacuum-tube cathode where electron emission occurs as a result of thermal motion, and photodiodes.
