No, you would need four transistors to get another 3db fall in noise contribution, then 8, then 16... law of diminishing returns, and in practice its not as good as the theory says too.
Two paralleled devices gives you the biggest improvement. Noise is uncorrelated between each device and so you take the square of each parallel noise source, add them together and then take the square root.
Two paralleled devices gives you the biggest improvement. Noise is uncorrelated between each device and so you take the square of each parallel noise source, add them together and then take the square root.
I wouldn't like to recommend any tbh because I'm not sure myself what would work best in that circuit. What did Doug use ? I've heard of good results sometimes being achieved with unlikely devices such as larger driver and medium power type transistors because of the larger die size. That was something I think J Linsley Hood found, remember reading it somewhere.
Given how they are paralleled in this circuit they will need to be very closely matched and track each other thermally for best results.
Analog devices makes the MAT14 array which might be of interest.. (good thermal tracking and relatively low noise)
MAT14 datasheet and product info | Matched Monolithic Quad Transistor | Matched Transistors | Analog Devices
I would recommend the MAT04 if you can find them, however it is discontinued.
The 2N4401/4403 are also very quiet and rather cheap, but thermal tracking may be an issue.
Analog devices makes the MAT14 array which might be of interest.. (good thermal tracking and relatively low noise)
MAT14 datasheet and product info | Matched Monolithic Quad Transistor | Matched Transistors | Analog Devices
I would recommend the MAT04 if you can find them, however it is discontinued.
The 2N4401/4403 are also very quiet and rather cheap, but thermal tracking may be an issue.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.