Post Go back to editing

AD8428: Noise at 2.5Mhz too high or normal?

Hallo everybody,

for our resistive sensor I designed a PCB with the AD8428.

To measure the Sensor I used a Bridge (I have 2 Sensors here, one for measuring and one as reference, to get rid of e.g. temperature dependencies)

The Sensor allows only small voltages, so I took an ADR420 and divided the voltage down, followed by a voltage follower (in this case I took an OP27, pretty old but we still got plenty of them)

I attached a photo of the noise I get when I connect both input pins directly (but it is the same noise I get when I connect my sensors or shorten the sensors).

I also attached the circuit and the layout (2 layers). Please note that in the attached layout I removed the GND polygon on top and bottom for better perceptibility.

I know that I should get a noisefloor at a level of: 1.5nV*sqrt(2.5MHz)*2000(Gain) * 3 (second OPAmp Gain)= 14mV RMS

But I get Peaks up to 100mV.

So, to finally come to my question, is this a normal noise or is it to much? Is there maybe a design failure?