Home Propeller Head Plaza

Technical and scientific discussion of amps, cables and other topics.

Re: Motsenbacher and Fitchen "Low Noise Electronic Design"

Low-Noise Design

I use a figure-of-merit of 5, meaning if the preamplifier has a gain of five then the noise contributed by the next stage is low enough to disregard. I got this from a Tektronix salesman who pointed out that if an oscilloscope has a bandwidth five times the signal bandwidth, then the scope signal amplitude is 0.1% accurate.
I use series-shunt feedback since this is the most stable. For example, if the input is a JFet, I pass the feedback to the source of the Fet. The secret of this amp is that the feedback is not applied to the input, so it has desirable features of both open-loop and feedback and is rock-solid.
In X-Ray preamps they always mount the JFet right next to the detector (microphone) and then run a cable to the rest of the amplifier.
I typically use the AD797 op amp if the input impedance is on the order if 1K or less. The AD797 input noise is only 50nV p-p between DC and 10Hz.
For a low-noise Fet amp, check out the Magnicon amps. They have very low noise Fet inputs - I would love to find out what Fet they are using. If you don't care about LF noise, then check out the MOXTEK (X-Ray Preamp) JFets, these are from Area 51. These will achieve 500MHz bandwidth at 1nV/Rt-Hz!
For affordable JFets see American Microsystems LSK170A.


This post is made possible by the generous support of people like you and our sponsors:
  McShane Design  


Follow Ups Full Thread
Follow Ups
  • Re: Motsenbacher and Fitchen "Low Noise Electronic Design" - PAMP 16:14:13 09/03/06 (0)


You can not post to an archived thread.