AD8475 using IN 0.8x for aditional low pass filter?

Hi,

I thought of using AD8475 as driver for 20 Bits 10KSPS...100KSPS SAR-ADC.

The signal frequency will be about 300Hz rectangle (chopped) and +/-10V peak.

I want to measure the DC-level with maximum possible accuracy.

1.)

Is this funnel-amp best suited for this application? Are there more precise / lower noise driver-ICs?

Current consumption / price / complexity doesn't matter.

2.)

I would like to implement an aditional low pass filter at the input. I will use the 0.4x inputs. So I thought to use

the 0.8x inputs for an additional low pass filter. Is this possible without significant degradation of low frequency parameters?

How to connect the capacitor(s)?

Would it even be possible to implement a two pole filter with that?

To get maximum DC-accuracy (and stability) I don't want an extra filter stage in front of funnel.

Probably this is interesting to other customers also.

Parents
  • 0
    •  Analog Employees 
    on Feb 10, 2014 7:15 PM

    Hello Alexxx,

    My apologies for the delay. What do you mean about the classification? Is that in reference to the new “Recognition and Awards Program”? I would be happy to forward any feedback you have to the team that developed it.

    I think AD8475 is a good choice for DC accurate applications. It is one of the few ADC drivers with integrated resistors, which significantly reduces the gain uncertainty. The 1/f noise is low and the offset and drift are very good for such a high speed amplifier.

    It is a good idea to put a capacitor across the 0.8x inputs when you are driving the 0.4x inputs. It gives you a free differential low-pass pole at about fp = 1/(2π*1.25k*C). However, this pole can vary ±20% due to the absolute resistor tolerance. I don’t know of a practical way to implement a second pole there, but you do get a second pole if you place another RC filter before the ADC inputs.

    I hope this helps.

    Best regards,

    Scott

Reply
  • 0
    •  Analog Employees 
    on Feb 10, 2014 7:15 PM

    Hello Alexxx,

    My apologies for the delay. What do you mean about the classification? Is that in reference to the new “Recognition and Awards Program”? I would be happy to forward any feedback you have to the team that developed it.

    I think AD8475 is a good choice for DC accurate applications. It is one of the few ADC drivers with integrated resistors, which significantly reduces the gain uncertainty. The 1/f noise is low and the offset and drift are very good for such a high speed amplifier.

    It is a good idea to put a capacitor across the 0.8x inputs when you are driving the 0.4x inputs. It gives you a free differential low-pass pole at about fp = 1/(2π*1.25k*C). However, this pole can vary ±20% due to the absolute resistor tolerance. I don’t know of a practical way to implement a second pole there, but you do get a second pole if you place another RC filter before the ADC inputs.

    I hope this helps.

    Best regards,

    Scott

Children
No Data