I am trying to find a capacitance to Digital converter for a range of 0 - 100 uF +/ 10 uF. I need a resolution of about 1 uF.

I took a look at the CN-0129 (http://www.analog.com/media/en/reference-design-documentation/reference-designs/CN0129.pdf) which shows how to extend the range of the AD7746 (originally 17 pF).

I worked through the example in CN-0129 and found that my FFix = 64,700,588, CDYN = 100,000 pF or 100 uF, FDYN = 7,324,219.

Using FDYN for my F (since it is the larger F value), and an R1 = 100 k, my R2 = 99.99997 k. This leaves me with a calculated 1.06 uF / LSB. Also note that I believe CN-0129 has an algebraic error where it suggests that R2 = R1 * (F + 1) / (F - 1), but actually R2 = R1 * (F - 1) / (F + 1) -> the + and - signs are flipped.

(I attached my calculations)

My questions:

Is extending the AD7746 to a range of 100 uF possible? in other words does this IC have a maximum extension and have I exceeded it with my desired range?

If I have exceeded the range of the ADD7746 is there an IC that can work in my larger range?

Thanks for any help that can be offered,

JW

Hi Jeremy,

The ADG7745/46 part was designed to be very sensitive for a small range of capacitance. I have never experienced anyone trying to scale it up so much as you need. The different between the resistors is tiny to realize such huge F factors. I feel it would be extremely difficult to control. We have only gone up to 300pF with the range extension circuit.

I will feed back your comment app note algebra here.

Regards,

Eric.