2
\$\begingroup\$

Here is a photo of the ADS1299 EVM from texas instruments. It is an 8-channel low noise analog front end chip. I understand that for each channel, the potential difference between plus and minus are measured by the AFE, but I am confused -- for 8 channels there are actually 32 pins on this EVM! I expected 16 (2x8 channels).
See J6 header to the left in the photo. Note that the two pins are joined by jumpers when not in use. The top set of 4 pins exposed on J6 are reflective of what the entire header would look like if I removed the jumpers.
I am clearly missing something.
enter image description here

\$\endgroup\$
5
\$\begingroup\$

One entire column of the pins are just ground. They exist for the use of the jumpers. The jumpers should be installed:

  • On the - pin of any channel where you want to make a single-ended reading instead of a differential reading
  • On both + and - pins of any channel which is not actively used.

The former is to set the non-inverting input to be at ground, thus making it essentially a single-ended ADC.

The latter is to tie both inputs of that channel to ground so they aren't a floating input. A floating input can potentially induce noise into the ADC affecting other readings on other channels. By tying the inputs to ground you are reducing that noise to a minimum.

\$\endgroup\$
  • 1
    \$\begingroup\$ Obviously, when not jumpered, those extra ground pins will be useful for tying the EVM ground to the signal source's ground, too. \$\endgroup\$ – Dave Tweed Jan 2 '15 at 16:56
  • 1
    \$\begingroup\$ Or for connecting a shield on the cable to should there be one. \$\endgroup\$ – Majenko Jan 2 '15 at 16:58
  • 2
    \$\begingroup\$ And if you use a ribbon cable with an IDC connector to make the connections to the EVM, you'll end up with a ground wire between any two signal wires, providing the maximum isolation for that kind of cable. \$\endgroup\$ – Dave Tweed Jan 2 '15 at 17:02
  • 2
    \$\begingroup\$ 1001 uses for a ground pin ;) \$\endgroup\$ – Majenko Jan 2 '15 at 17:03
0
\$\begingroup\$

Based on the PCB I think that the leftmost row of pins on J6 (odd pins in 5-36) are not connected to anything. If this is true I am not sure why TI user a two row header for J6. Some but not all the left J6 pins appear to be connected to something, but not sure what yet (pdf wasnt great quality to zoom in)enter image description hereenter image description here enter image description here

\$\endgroup\$
  • 2
    \$\begingroup\$ It looks like at least half of the extra pins are connected to ground, making it easy to feed in unbalanced signals by tying the (-) input to ground with a jumper. \$\endgroup\$ – Dave Tweed Jan 2 '15 at 15:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.