AnsweredAssumed Answered

AD8318 oscillation

Question asked by willplatts on Feb 18, 2016
Latest reply on Feb 23, 2016 by willplatts



I have encountered a problem with the AD8318 logarithmic amplifier. We are using the AD8318 in simple measurement mode (i.e. VSET tied to VOUT). We are enabling/disabling the device via the ENABLE pin connected to a microcontroller (so driving the pin from 0V to 3.3V). In some devices, when the device is enabled, the output of the AD8318 goes into oscillation (see attached screenshot from oscilloscope).


I have done some investigation, and found the following:


  • The instability only seems to happen when we use a discharge resistor on the VOUT pin. I am using this to speed up the falling edge, and also to overcome a problem where high power signals cause a strange waveform on the output of the log amp.
  • If the discharge resistor is 160R, it happens almost every time the enable pin is taken high (only on some devices though)
  • If the discharge resistor value is increased, to say 500R, 750R, 1k, the problem happens less often but still sometimes.
  • Touching a scope probe to the output of the log amp stops the oscillation (attached capture was taken with a high impedance/low capacitance FET probe)
  • The oscillation is well beyond the normal output range of the log amp: as you can see, it is swinging all the way to 0V, and up to around 4.5V. Normally the output is only from around 2V to 0.5V
  • Adding extra decoupling capacitors makes no difference.


Does anyone have any ideas what is causing this problem, or have seen anything like it?


Thanks in advance


Will Platts

EA Technology