Post Go back to editing

ADA4800 - Graph of Output Voltage Noise

  Good afternoon.  I have a customer considering the use of the ADA4800 (Low Power, High Speed CCD Buffer Amplifier) with one of the ADI AFEs.  The data sheet only indicates the "typical" output voltage noise at a frequency of 20MHz.  Do you have a graph showing a more complete picture of this parameter over then entire 400MHz potential bandwidth that I can send to the customer?

  Thank you.


  • Hi Rich,

    Apologies for the delayed response. Unfortunately, we don't have data/graph showing the output voltage noise density for the entire bandwidth(400 MHz). Spectral noise density is usually composed of 1/f noise and white noise. If you're concerned about white noise, typically it does not vary with frequency and is usually equal until the bandwidth rolls it off. It is likely that at 400MHz, the noise spectral density of ADA4800 is still roughly 1.5nV/rtHz. This is different from 1/f noise which is a low frequency noise and has a voltage noise spectral density that is inversely proportional to the square root of frequency. It also has a spectral density that rises at 3 dB/octave and is usually specified at the corner frequency of 1 to 10 Hz.

    You may want to read these articles discussing Op Amp Noise:

    We also have a very good video discussing how to convert the spectral noise density to RMS:



  • Hello Emman,

    We're also looking at the ADA4800 for use with scientific grade CCDs and have a few questions.  The first concern is with the current noise of the current source load.  Current noise on the load will develop a noise voltage across the CCD's on-chip amplifier output impedance.  That noise will then be buffered by the ADA4800 internal op-amp and show up on the output.  Is there any information on the noise current?  I didn't see any on the data sheet.  It seems to us that it is somewhat noisier than "discrete" loads we have been using, but we don't have fully quantified measured results to share (at least not yet).

    The second issue is the allowable input current that can safely be input to the input pin during power-up.  The absolute maximum ratings only state Vcc is the maximum.  Are there protection doides built in?  How much current can they handle safely?  We use CCDs that have output voltages > 20V, but we can put a dropping resistor, with a bypass capacitor across it, to reduce the voltage at the input to < Vcc.  But during the power-up cycle, the voltage might try to go above Vcc until the capacitor is charged.  Do you have any experience or guidelines for this situation?  We might be able to control the rise of the CCD Vdd so the capacitor gets to equilibrium, if we know how much current we had available.  Most i.c. protection networks can handle 10 ma or more at least for a short time.



  • Emman,

    I would like to add my recommendation that ADI add the noise spectral density graphs to the data sheet.  We tend to use CCDs over a wide range of operating speeds (slow scan for maximum sensitivity, higher speeds for more dynamic imaging).  That was one of the appeals of the ADA4800 -- it seems like it will work over a wide speed range.  But to really predict performance over a wide range, we would need the noise spectra density or equivalent information. 



  • This question has been assumed as answered either offline via email or with a multi-part answer. This question has now been closed out. If you have an inquiry related to this topic please post a new question in the applicable product forum.

    Thank you,
    EZ Admin