ads42lb69 with DC-coupled Input appears to have an impulse response

Calvin Field

Other Parts Discussed in Thread: ADS42LB69, ADS62P49

I have an ads42LB69 that I dc-couple through a differential amplifier. I see an exaggerated response when I apply a square wave, there is no ringing just 75mV of overshoot followed by a 2us decay. I don't have any distortion issues with sine wave input (good SFDR and SNR). I am able to probe the input to the ADC and do not see this amount of overshoot/decay. I have bypassed the amplifier and driven the ADC direct single ended (offset to VCM) and see the same response. I ran the same experiment on an ADS62P49 and do not see the problem.

over 9 years ago

0 Ebenezer Dwobeng over 9 years ago

TI__Genius 9695 points

Hi Calvin,

The overshoot is likely due to the step response of the internal buffer at the analog input of ADS42LB69. I will find out more and get back to you. Can you give me an idea what is the slew rate and frequency of your input signal?

Thanks,

Eben.

0 Calvin Field over 9 years ago in reply to Ebenezer Dwobeng

Prodigy 180 points

The slew rate is ~200 mV/ns on a 1.6 Vpp 100 kHz 50% squarewave.

Regards,

Calvin

0 Calvin Field over 9 years ago in reply to Ebenezer Dwobeng

Prodigy 180 points

Do you have any more information? This problem is critical and we are in production stop work. Thanks.

0 Ebenezer Dwobeng over 9 years ago in reply to Calvin Field

TI__Genius 9695 points

Calvin,

The parasitic inductance of the package is contributing to this overshoot and probing the signal on the outside of the package will not reveal this overshoot. A small series resistor (R8,R12) is recommended at the analog input as shown in the figure below to minimize this. I will recommend using a resistance of 15ohms for R8,R12. You may use a higher series resistance (20ohms or higher) to further reduce the overshoot but at the cost of lower input bandwidth and signal attenuation.

Thanks,

Eben.

0 Calvin Field over 9 years ago in reply to Ebenezer Dwobeng

Prodigy 180 points

I've tried all sorts of matching networks including one similar to your suggestion (33.2 Ohms, 6.8pF). I cannot get the shape to change at all. There is something really peculiar about this response. Is your recommendation based on actual experiment (running a square wave into a dc-coupled ads42lb69) or is it based on simulation?

0 Ebenezer Dwobeng over 9 years ago in reply to Calvin Field

TI__Genius 9695 points

Calvin,

The recommendations were from design and I have verified on the bench that it does not resolve this particular issue (see screenshot below). If you can give me your email address, I will like to work with you on this offline.

Thanks,

Eben.

0 Masaharu Katayama over 8 years ago in reply to Ebenezer Dwobeng

Expert 1420 points

Hello Eben,

I met similar critical issue. Below is waveform and circuit.

100KHz rectangle input, and Digital gain is 2.5VppFS.

Some variations of series C and R values are changed but a sag wave does not changed.

Which block in ADC causes this? and is there a solution?

Here is my email (if need off-line), please respond ASAP

katayama-m@clv.macnica.co.jp

Best regards,

Masa

0 jim s over 8 years ago in reply to Masaharu Katayama

TI__Guru* 91947 points

Masa,

I will reply off line on this.

Regards,

Jim

Data converters

Data converters forum

ads42lb69 with DC-coupled Input appears to have an impulse response