Question to Correct Usage of AD8422

Hi pre ,

Good day. Can you share more details about your sensor, such as the type of sensor, output range, and output connections, so we can suggest a better setup? As for the offset, this should not be an issue since the AD8422 offers an offset voltage at around a hundred microvolts. Also, can you share what version of LTSpice did you used? Lastly, I recommend the usage of guard rings for the input signal. You can refer to this article Layout For Precision Op Amps. This should apply to any op-amps including AD8422. Thank you.

Regards,
Gilbeys

Dear Gilbeys,

first, thank you for your fast reply.

The sensor is a self-designed short antenna (on a PCB). The purpose is to measures low-frequency (1 Hz ... 20 kHz) electric fields. It is basically a piece of wire (30 pF coupling capacitance) with a 10 MOhm coupling resistance towards ground. The measured voltage towards ground is then put through a simple low pass filter (5 MOhm and 60 pF). It is connected via an u.fl cable to the AD8422 (I neglected this small additional capacitance of a few pF in the shematics above). Let me know, if I can provide further information.

I use LTSpice Version (x64): 24.0.9

Also, I will check you guard ring reference and may come back with questions.

Best wishes!

Hi pre ,

I've done some analysis on your circuit focusing on the AD8422. It seems that the offset you see in the simulation is due to the impedance seen by the amplifier at the input. From my calculation, the impedance seen by the +IN is around 58.7 MHz. The input offset current of AD8422 is 0.5 nA. The offset voltage for this should be around 29 mV which could explain the offset in the simulation. To resolve this, I would recommend adding an equivalent resistance of 58.7MHz at the non-inverting terminal. I just want to know the reason behind the choice of AD8422 as the buffer amplifier for your design. Thank you.

Regards,

Gilbeys

Dear GLADION ,

thank you for your answer. I guess with 58.7 MHz you meant to write 58.7 MOhm, can you confirm?

We chose the AD8422 because we need a high input impedance (it should be much higher than our antenna impedance, hence we cannot directly go into the PGA); it has a very low power consumption; we can set an initial gain; it has sufficient GBP and can be powered by +-5 V. Also, as we measure harmonics, it has a good noise density and low harmonic distortion, to my understanding. Coming back to my initial question 1.: Do you think the AD8422 is the right choice or would you use/recommend another amplifier as voltage follower/input buffer?

Thank you!

Hi pre ,

Sorry. Yes, that's correct. That is 58.7 MOhm. To answer your question, your setup should work fine. However, you can also design the circuit using a PGIA. LTC6373 should be able to perform well with your sensor and could substitute the three components in your schematic: AD8422, LTC6910-2, and AD8476. I've attached a schematic simulation which you could use as a reference. Pre_EZone.asc

LTC6373 could also drive your ADC without some issue.

Thank you.

Regards,
Gilbeys

Hi pre ,

Sorry. Yes, that's correct. That is 58.7 MOhm. To answer your question, your setup should work fine. However, you can also design the circuit using a PGIA. LTC6373 should be able to perform well with your sensor and could substitute the three components in your schematic: AD8422, LTC6910-2, and AD8476. I've attached a schematic simulation which you could use as a reference. Pre_EZone.asc

LTC6373 could also drive your ADC without some issue.

Thank you.

Regards,
Gilbeys

Hey GLADION ,

thank you for the confirmation, this really helps. Actually, we have used the LTC6373 before and in worked well. The only, unresolvable, downside was the compartively large sonsumption, so we had to come up with an alternative design.

I will continue using the proposed setup, but consider including guard rings and DC bias resistance!

Thanks alot!