Hi,
I would like to know benefit of Mux-Friendly.
Is benefit of Mux-Friendly to shrink a settling time as below?
Below waveforms are mentioned on the datasheet page 1.
Best Regards,
Kuramochi
Kuramochi-san,
Indeed the MUX friendly design achieves faster settling time than more conventional CMOS op amps using back-to-back input diode clamps. That is accomplished by eliminating the diodes and high surge currents that can occur when a MUX preceding the conventional op amp has widely different voltages from one channel to the next.
If you would like more details about how exactly the MUX friendly op amp design achieves what it does see the following TI TechNotes on the subject:
If you have any questions after reading the TechNotes let us know.
Regards, Thomas
Precision Amplifiers Applications Engineering
Hi Kuramochi,
I did run a simulation with the OPA189. Assume a typical multiplexer application with a Ron of 200R and a transition time of 100ns. The input signal at the one input of multiplexer shall be +10V and the input signal at the neighbour channel shall be -10V. Then the following will happen when the multiplexer is switching between these two channels:
You see that Vout shows a smooth rise and needs about 800ns to settle. Also, through the simulated Ron of multiplexer (R1) almost no current is flowing.
Then I have added back-to-back diodes between the inputs of OPA189 to simulate an OPAmp with protection diodes between its inputs:
A huge difference can be seen: As the output of OPAmp cannot immediately follow the input signal during the transition, the input signal sees a short circuit into the output of OPAmp! As consequence a huge current is flowing through the simulated Ron of multiplexer (R1). This short circuit current has not only to be delivered by the source driving the input of multiplexer, but has also to be handled by the OPAmp itself. As consequence the output stage needs an additional amount of time to recover from this short circuit event. So, Vout of OPAmp needs about 900ns to settle.
Kai
