Title: AD9213 Output Square Wave Exhibits Charging/Discharging During High and Low Levels

Hi ITRIHW 

Thanks for using AD9213.

Your observations are normal and expected for the AD9213 balun‑based input path. The charging/discharging effect is not a problem with the ADC itself but a consequence of AC coupling and transformer behavior.

• When the ADC input is driven through a balun and AC‑coupling capacitors, the front end effectively behaves as a high‑pass filter. A 10 MHz square wave contains DC content and low‑frequency harmonics, and these components are significantly attenuated as they pass through the balun and AC‑coupling network, causing waveform distortion and baseline tilt.
• A better approach is to use a wideband differential amplifier to drive the AD9213, especially if you need to preserve the waveform shape. The amplifier can convert the single‑ended square wave into a proper differential signal for the ADC. Just make sure to use DC coupling and remove both the balun and the AC‑coupling capacitors, since these elements would otherwise filter out the low‑frequency components you want to maintain.

See AN-2539: Single-Supply, DC-Coupled 16-Bit, 125 MSPS Analog Front End for Bipolar Input Signals | Analog Devices for your reference.

Hi JAlipio 

Thank you for the clarification and explanation.

We understand that the observed behavior is due to the balun and AC-coupling network acting as a high-pass filter.

Regarding AN-2539, the amplifier ADA4930-1 used in that reference design provides approximately 1.3 GHz small-signal bandwidth,

but its linear performance is typically optimized for applications below a few hundred MHz.

Since AD9213 operates at GSPS rates and is typically used for multi-GHz input signals, 

could you please advise whether there is a recommended wide-band differential amplifier solution (GHz range) suitable for DC coupled operation with AD9213?

Specifically, we are looking for a driver that:

• Supports multi-GHz bandwidth

• Allows DC coupling

• Is compatible with AD9213 input common-mode requirements

Any reference design or recommended part would be greatly appreciated.

Thank you for your support.

Hi JAlipio ,

Sorry to bother you.

Just a kind follow-up regarding our question about a wide-band DC-coupled driver for AD9213. When you have a moment, we would sincerely

appreciate any update you may be able to share.

Thank you very much for your continued support.

Hi ITRIHW 

For DC‑coupled, multi‑GHz‑bandwidth front‑end design for the AD9213, the ADMX6001 platform serves as a strong reference point. Its documentation provides practical examples of broadband, DC‑coupled analog interfaces, GHz‑range signal‑chain implementations, and common‑mode biasing and layout techniques appropriate for high‑speed ADC systems.

While the ADMX6001 is not an ADC‑driver IC, its front‑end architecture, biasing methods, and wideband amplifier stages offer design approaches that are directly applicable when creating a custom DC‑coupled differential interface for the AD9213.

ADMX6001 Reference Materials:

• User Guide: UG-2361 (Rev.A)
• Evaluation Hardware Overview: EVAL-ADMX6001 Evaluation Board | Analog Devices
• ADMX6001 Wiki Resources: ADMX6001-EBZ: DC-Coupled 10GSPS Digitizer Evaluation Board [Analog Devices Wiki]

If you require access to the ADMX6001 evaluation hardware or additional technical information and design files, you may also request support through your local ADI FAE, who can help facilitate this.

Hi JAlipio ,

Thank you for the information and for pointing us to the ADMX6001 reference materials.

We are currently using ADL5505 connected at the AD9213 output.

After passing through the driver amplifier stage on the board, we observed that the overshooting seen in the input waveform disappears in the output waveform.

10 MHz Signal from SG: 

AD9213 with ADL5505 connected – 10 MHz output waveform

Could you please help us understand whether this behavior is expected?

If possible, any insight regarding the signal conditioning or bandwidth characteristics of the driver stage would be very helpful for our analysis.

Thank you very much for your support.