THS4503EVM: [THS4503EVM] Issue when using non-inverting configuration

It looks like the image did not go through in the original post.. adding it here as an attachment.

FDA's require an input resistor, think of them as a pair of inverting op amps with an added output CM control loop,

Thanks for the quick reply. So instead of removing R17, it should be replaced with a ~50 ohm resistor?

David

So the whole thing of setting R values for single to diff gets amazingly complicated for the FDA. In fact, the full closed form solutions for a matched input impedance did not enter the public domain until I worked them out and published them back around 2014. Before then (and still on the ADI material) TI and ADI and NSM used an iterative approach. 

The THS4503 was done before those solutions were there - if you want to see what those look like, I first put them into a TI datasheet for the THS4541 release. Look at page 37 of that datasheet. I originally developed them to support the ISL55210 release - still one of the lowest input noise FDA's. 

So to go further for you, 

1. what is your source impedance

2. Do you need to match that impedance at the FDA input

3. what single to diff gain is desired.

4. 

Well David, 

You had said you had this resolved, but I had some time this AM and need to exercise my FDA design tools occasionally. Yes, the standard EVM set up is just slightly off on their input R values - this was before the active impedance aspect of single to diff FDA was well understood.

Here is a review of the Fig1-1 in that EVM document. Review of the THS4503 EVM schematic.docx