ADS124S08: ADS124S08 with 4x 3wire RTD input

Hi Manu,

I think there may have been a typo and I believe the TI Designs you actually are referring to is the TIDA-00110.  Please see my responses below.

Best regards,

Bob B

Manu Abraham said:

Part Number: ADS124S08

Hi,

Referencing TIDA-0010 for connecting 4x 3 wire RTD sensors

came up with the following schematic:

In this regard, I have the following questions:

1. Is a switch (TS3A5017) required for REFP/N, much like IEXC1/2; but isn't the current path complete only with IEXC1/2 ? A single switch alone suffice as in the attached schematic, if that's the case ? [Bob] No switch is required for REFP/N.  The current is directed to the RTD via the IDAC switch.  For RTDs not selected, the circuit is open and no current will flow.  So a single switch can be used for directing the IDAC current and no other switch is required.

2. IN1/2 control using ADC GPIO's: Any issues in doing so ? [Bob] The issue as I see is the switch is powered from 3.3V and the analog is powered by 5V.  The GPIO output is coming from the analog supply so its logic will be 5V.  The TI Design uses a 3.3V analog supply voltage.  You may want to consider reducing the voltage to 3.3V for the AVDD supply of the ADS124S08, or find a different switch capable of 5V operation.

3. AINCOM should be left floating, since the measurement is not single ended ? [Bob] AINCOM is really no different than any other input.  Its labeling just makes it easier to reference when connecting to single-ended or pseudo-differential inputs.  If this input is not being used it can be left floating.

4. Should the internal reference be used, or an external reference such as a REF6225/REF5025 be used with REFP0/N0 as used in sbau272a.pdf ? [Bob] To get the best noise and drift performance a ratiometric measurement should be used where the excitation source for the RTD and reference are the same.  If an external voltage reference is used, then any IDAC drift will show as error in the measurement.  Using a ratiometric measurement the conversion result is simply a ratio of the RTD resistance to the reference resistor resistance. This method is discussed in section 10.2 of the ADS124S08 datasheet.

5. Is a schottky diode really needed on each of the RTD inputs, ie on the reference signal from IDAC1/2 ? [Bob] What level of input circuit protection is up to you and the protection can be accomplished in a variety of ways.  The Schottky diode prevents external sources (EMI/RFI) from driving current back into the ADS124S08.  Some people use resistors to limit current, while others use nothing.  It really depends on your system requirements.  The TI Design shows numerous methods of protection including various types of transient protection on the inputs.

Thanks,

Manu

Hi Manu,

I would suggest you review this document:

http://www.ti.com/lit/an/sbaa275/sbaa275.pdf

You have switched from a low-side reference to a high-side reference.  This can work, but for 3-wire measurements the current in the pathway changes and there are some implications. 

The jumpers you show in a row can be avoided by not turning on the IDAC through that path when a 2-wire device is connected (software versus hardware).  The jumpers shown as a column of devices connect to the wrong side of the Rbias resistor placing the 2-wire measurement at ground outside of the PGA input range.

Best regards,

Bob B