AD4007 / AD4003 - TIA interface

Hi Thorsten,

I'm looking into this and get back to you asap.

regards,

LLoben

Thx !

Hi Thorsten,

I'd like to get more information about your application.What's driving the need(performance- SNR/THD, Linearity?) to use a differential SAR ADC?.What is your input signal bandwidth. I presume its below 100Hz. Although we haven't verified the AD400x family performance with AD8597 as a driver,it should still be able to drive the AD4000/03/07 and achieve decent performance for low frequency input signal(<=100Hz) with High-Z mode enabled. I'd suggest you to use the ADA4940 as a single-ended to differential driver between the AD8597 and AD4003/07. The AD400x eval board (user guide) has a footprint for the AD4940,which will allow you to evaluate the performance for this signal chain.See this article if it helps. Unlike traditional SAR ADCs, a unique combination of high-input impedance mode,reduced input current and longer acquisition phase reduces the settling requirements on the ADC driver and hence a higher noise and /or lower power/bandwidth amplifier can be used.

regards,

LLoben

Thank you for the feedback, LLoben !

The application basically is scanning with a step frequency of 200kHz. The settlement time of the stimulus (input of the photodiode) is pretty fast (<300ns). For the photodiode/TIA side simulation/calculation based on the AD8597 let me expect that we will achieve a settlement time to <0.01% of less than 3us. Within the 5us timing window (because of the 200kHz scanning frequency) there is enough time left after settlement for conversion with a SAR ADC like the AD4007.

Based on a two stage TIA-configuration with the AD8597 for a photodiode current of 1mA an SNR towards 100dB is achievable - which corresponds well to the achievable SNR of the ADC and the requirements of at least 92dB SNR.

Hope this gives at least a short intro to the boundary conditions of the application.

If there should be questions please let me know ...

As you said you would assume a "decent" performance of the AD8597 with the AD4007 at signal frequencies of about 100Hz - my question would be : What is your assumption with regard to the application case described above -obviously thew performance requitrements are "slightly" different ...  My assumption that directed me towards the AD4007 was that because of the High-Z-Mode it should be "easy" for the AD8597 to settle in the available time... How is your perception ?

Thx for the hint towards the AD4940 - I will look into a combination with the AD8597 as a TIA ...

Best, Thorsten

Hi Thorsten,

Have you looked at Analog Devices photo diode tool for this TIA configuration?

http://www.analog.com/designtools/en/photodiode/

The AD8597 should be able to achieve decent performance based on its datasheet specs (10MHz GBP, 1.1 nV/√Hz noise) and what we have seen with other precision amplifiers (See figures 45 and 46 in datasheet) with High-Z enabled. However, I’d encourage you to evaluate this signal chain performance to make sure it meets your requirements.

Regards,

LLoben