ADS124S06: Temperature drift matching between IDACs

Hi Ohno-san,

There are two factors to consider.  One is the initial matching and the second is the drift matching.  There are two graphs in the ADS124S06 datasheet that show the characteristics on page 20 in Figures 28 and 29.  The main factors with the drift is the reference stability by which the IDACs are derived and the current sources themselves.

One way of canceling the effects of IDAC matching/drift is by swapping IDACs in the measurement.  An example on how to do this is show in A Basic Guide to RTD Measurements on page 6 in section 1.4.2.

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

Best regards,

Bob B

Hi Ohno-san,

I'm don't believe I have more to add to my previous comments, but perhaps I need to clarify.

Best regards,

Bob B

Kenji Ohno1 said:

Hi Teams,

1. I have doubts on Temperature drift matching between IDACs in data sheet P11.

For example, when IDAC1 and IDAC2 are 500uA,

IDAC1 = Xppm / ℃

IDAC2 = Yppm / ℃

Is it correct to guarantee that X-Y = typ.2ppm / ℃ and max.15ppm / ℃? [Bob] The specification is very general relative to a wide range of currents from 250uA up to 2mA.  Using the information in Figure 29 you can track the response of IDAC mismatch specifically to the actual output current.  I have included a copy below:

The above graph equates to the typical response of 2ppm/deg C of full-scale (500uA) across the operating temperature of the device.  It is possible that the maximum drift could be 15ppm/deg C, but the trend would be similar.

 ①When X = 5ppm / ℃, Y range is 3 ~ 7ppm / ℃ in TYP value [Bob] Use of the IDACs are meant to be used for excitation in a ratiometric measurement which results in an output code as a ratio of the reference resistor (for RTD measurement) independent of a single current drift.  It is the mismatch of the current between the two IDACS when used in a 3-wire RTD measurement that should be considered and not the total drift of the IDAC.  So you would only consider the mismatch of 2ppm.

 ②When X = 120ppm / ℃, the range of Y is MAX value 105 ~ 120ppm / ℃

 * Since the temperature drift of IDAC is 120ppm / ℃ at the maximum value, Y≤120ppm / ℃ [Bob]  Similar to what I just stated, it is only the mismatch of the current that can affect the result and not the total current output of the IDAC itself as the measurement is ratiometric.  Chopping, or swapping the IDAC current routing and taking an average of the measurements removes the mismatch error.  This method is described in the application referenced in my first response.

2. Temperature drift (each IDAC)

・ What is the main factor of temperature drift?  [Bob] As I stated in my first response, the factors affecting the drift are the internal reference drift and IDAC device construction as these devices will change response over temperature.  The internal reference is used as the reference source for deriving the IDAC current.

・ Do you have the characteristic data in the range of Ta = 50 ℃～90 ℃? [Bob] Again I refer you to the graph above (Figure 29). You can see that the response for 500uA between 50 and 90 deg C changes about 0.01 %.

　In this range, will the temperature drift either positive or negative with respect to the rise in ambient temperature? [Bob] Again, consult the graph for drift behavior.

Best Regards,

Kenji