AD8338 stability issues

Hello guimpt,

Thanks for reaching out. Can you provide me the following:
1. Amplitude of the signal entering the INPR and INMR pins. If you can provide some scope shots, it would be better.
2. Target output signal level.
3. Output scope shots taken at the OUTP and OUTM pins describing the issue.

In parallel to this, can you also try the following:
1. Temporarily remove C37 and C38 and check if the issue still persists.
2. Use other probes to get the output signal. Sometimes, oscilloscope probes introduce unwanted capacitances and cause capacitive loading that can introduce instability. 

Regards,
Paul

Hello  ,

Thank you for your reply. Please find the requested information below:

1. Amplitude of the signal entering the INPR and INMR pins: The AD8338 is operated across its full 80 dB gain range. At maximum gain, the intended signal level at the input is in the low-microvolt RMS range, while at minimum gain the input can reach tens to hundreds of millivolts differential.

2. Target output signal level: Approximately 0.1 Vrms differential.

3. Output scope shots taken at the OUTP and OUTM pins describing the issue. Attached below is an example of the issue. The yellow trace shows the VGA output (OUTP/OUTM, differential), and the blue trace shows a disturbance injected into the antenna using a second antenna. When the disturbance causes output saturation, the AD8338 enters a sustained oscillation, which persists even after the disturbance is removed.

We have tried both of your suggestions:

• Temporarily removing C37 and C38
• Using different oscilloscope probes

In both cases, the same behavior is still observed.

Please let us know if additional measurements or information would be helpful.

Hello guimpt,

From what I've observed, by the time the disturbance has been injected, it forces the op amp to output about 3Vpp (or 1.5Vpeak) differential. AD8338's differential output voltage range is only guaranteed up to 1.4Vpeak or 2.8Vpp and exceeding this range will more likely drive the op amp to operate non-linearly. This could be the reason why your output voltage signal oscillates and continues even if the disturbance has been removed. We cannot guarantee AD8338's performance outside its guaranteed operating conditions as indicated on its data sheet.

As a recommendation, I would suggest filtering the disturbance to reduce its level or reducing the gain to make sure your differential output voltage signal level is within the differential output voltage range specified in the part's data sheet.

Hello guimpt,

From what I've observed, by the time the disturbance has been injected, it forces the op amp to output about 3Vpp (or 1.5Vpeak) differential. AD8338's differential output voltage range is only guaranteed up to 1.4Vpeak or 2.8Vpp and exceeding this range will more likely drive the op amp to operate non-linearly. This could be the reason why your output voltage signal oscillates and continues even if the disturbance has been removed. We cannot guarantee AD8338's performance outside its guaranteed operating conditions as indicated on its data sheet.

As a recommendation, I would suggest filtering the disturbance to reduce its level or reducing the gain to make sure your differential output voltage signal level is within the differential output voltage range specified in the part's data sheet.