• Locked Locked
  • Replies 2 replies
  • Subscribers 46 subscribers
  • Views 2376 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Leakage current for SAR ADC input channels

Leo Costello
Prodigy 150 points
Other Parts Discussed in Thread: ADS7841, ADS7828, ADS7844

BACKGROUND:
This question is about the leakage current into the ADC input channel(s) of SAR ADC's such as the ADS7844, ADS7841, ADS7828 or many other TI ADCs in this class.

The datasheets provide only a "Typical" leakage current spec (usually +/- 1uA ), but do not provide "Min" and "Max" values for leakage current, or temperature drift specs. We are designing an improvement for a legacy product that was designed without considering this leakage current. This ADC (the ADS7828) measures several DC voltages, but all have very high source resistances (~1 Mohm), so clearly the leakage current MUST be considered for accurate measurement. If we knew the Min and Max values of leakage we could determine what the maximum error will be and proceed accordingly. Having only the "Typical" value doesn't provide enough information for our application. We would prefer not to have to perform our own "device characterization" by testing, for example, a batch of 100 or more units to empirically predict the Min and Max leakage currents.

SPECIFIC QUESTIONS:

1) Does TI have additional device characterization data for the range of leakage currents we can expect to see in production quantities, and over the device operating temperature range?

2) We have measured the leakage current on a few devices and found much lower levels than the typical spec (0.300uA to 0.500uA on 5 devices). Can TI indicate whether the +/- 1uA typical spec is a conservative number and the vast majority of devices will have much lower leakage currents?

3) Is there anybody who has designed-in this class of capacitive SAR ADC's who can share more information on the leakage current range?

4) Since it appears that this leakage current is essentially the input offset current of the Track and Hold (Sample and Hold) op amp, it makes sense that the leakage current can be either positive or negative, as indicated by the "+/- 1 uA" specification. Could someone please confirm that for production quantities, the leakage current on a given device can be either positive or negative?

Thanks very much!

Leo

 

 

 

 

 

  • TI__Guru**** 180961 points

    Hi Leo,

    I'll check around and see what sort of characterization data I can get you for the leakage current of the ADS7828.  The +/-1uA is conservative - that number comes from characterization data with added guard band.  Since it is listed as 'typical' this usually means the leakage is not production tested.  Note that the ADS7841 data sheet lists the leakage as 200nA MAX - that implies leakage is production tested on all parts.

    The leakage currents can be +/- and there are contributions possible from the S/H amplifier as well as the ESD cells.  One side note - typically SAR converters are driven from a low impedance source and more often than not, we recommend that you use some sort of buffer ahead of the ADC input.  This often provides the best, most consistent performance.  Here are a few articles that you might find useful:

    Designing SAR ADC drive circuitry, Part I: A detailed look at SAR ADC operation. Analog Zone. Download at: http://www.analogzone.com/acqt0221.pdf

    Designing SAR ADC drive circuitry, Part II: Input behavior of SAR ADCs. Analog Zone. Download at: http://www.analogzone.com/acqt1003.pdf

    Designing SAR ADC drive circuitry, Part III: Designing the optimal input drive circuit for SAR ADCs. Analog Zone. Download at: http://www.analogzone.com/acqt0312.pdf

  • in reply to Tom Hendrick
    Prodigy 150 points

    Hi Tom,

    Thanks very much for your reply -- it is a beginning in helping me deal with the problem I have been handed in this legacy design (i.e., a design I inherited, and with thousands of units in the field needing a fix!).

    Before writing my original post, I did some spot-checking of the leakage current spec on several of the ADS78xx ADC family devices, including the ADS7841, so your "200nA MAX" comment puzzled me. I went back and checked, and it is only the automotive catalog "-Q1" version that specs a 200nA MAX leakage current. The standard ADS7841 specs the usual "+/-1 uA Typ" figure. And I find it interesting that the -Q1 versions apparently specs a positive-only leakage current -- in other words it specs " 200 nA MAX" and NOT "+/- 200 nA MAX."  So, it appears that the -Q1 part is screened both for magnitude AND polarity of the leakage current.

    Or, is the proper interpretation of the datasheet that the unspecified MIN leakage current could be -200 nA or lower (that is, a higher magnitude but negative)??

    Thank you for the links to the SAR ADC app notes. I have studied the TI app notes on this subject, and so am familiar with the signal conditioning (either op amp or ~0.12uF cap) needed for capacitive SAR inputs, even when measuring DC voltages. But these 3 papers go into even greater detail -- so thanks again.

    Leo

     

     

    Tom Hendrick said:

    Hi Leo,

    I'll check around and see what sort of characterization data I can get you for the leakage current of the ADS7828.  The +/-1uA is conservative - that number comes from characterization data with added guard band.  Since it is listed as 'typical' this usually means the leakage is not production tested.  Note that the ADS7841 data sheet lists the leakage as 200nA MAX - that implies leakage is production tested on all parts.

    The leakage currents can be +/- and there are contributions possible from the S/H amplifier as well as the ESD cells.  One side note - typically SAR converters are driven from a low impedance source and more often than not, we recommend that you use some sort of buffer ahead of the ADC input.  This often provides the best, most consistent performance.  Here are a few articles that you might find useful:

    Designing SAR ADC drive circuitry, Part I: A detailed look at SAR ADC operation. Analog Zone. Download at: http://www.analogzone.com/acqt0221.pdf
    Designing SAR ADC drive circuitry, Part II: Input behavior of SAR ADCs. Analog Zone. Download at: http://www.analogzone.com/acqt1003.pdf
    Designing SAR ADC drive circuitry, Part III: Designing the optimal input drive circuit for SAR ADCs. Analog Zone. Download at: pdfhttp://www.analogzone.com/acqt0312.