• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 45 subscribers
  • Views 1615 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.

ADS1220: Determine ENOB for a given Vref and GAIN

Pablo G.C
Prodigy 40 points
Part Number: ADS1220
Other Parts Discussed in Thread: ADS122U04, ADS122C04, ADS124S06, ADS124S08EVM

Hi everyone,

I want to use an ADS1220 to measure two PT100 as described in the first page of this appnote . The datasheet of the device specifies ENOB based on Gain and datarate for 3.3V AVDD and internal reference but since I'm using the IDAC to feed Rref I don't know if I can trust this tables.

What I need is 45000+ counts from a PT100 in the 0-45C Range (100 to 117.47Ohm) but as I've said I'm missing the ENOB to verify if can get that many counts, can someone lend me a hand? Thanks

  • 0
    TI__Guru** 113765 points

    Hi user5352065,

    Welcome to the E2E forum! Typically ENOB is not a good unit of measure for determining the outcome because this value directly relates to the value of one LSB of resolution and the LSB value will change depending on reference.  With an RTD ratiometric measurement where the excitation is the same for both the reference and the RTD, the conversion result is based on the ratio of the RTD value to the reference resistor value.  The ability of maintaining an accurate measurement is also dependent on the stability of the reference resistor.  I would suggest looking at the http://www.ti.com/tool/TIDA-01526 which uses the same concept, but with design information for the ADS122U04/ADS122C04.  The ADS1220 has a very similar analog input structure so much of the discussion applies.  In particular look at section 2.4.

    Best regards,

    Bob B

  • 0 in reply to Bob Benjamin
    TI__Mastermind 23495 points

    Hi user5352065,

    I ran some calculations using the circuit implementation as shown in our TechNote.
    Unfortunately it seems like that the ADS1220 will not be able to achieve the noise performance you are looking for. Assuming that you don't care about power consumption, I configured the circuit for IDAC=1.5mA, Gain=4, Data Rate=20SPS and RREF=600Ohm. With those settings I get to a temperature resolution of about 1.6mK to 2mK, while you are asking for sub 1mK resolution I believe. This calculation assumes rms noise values. In case you are asking for 45,000+ noise-free counts, this will make the situation even more challenging as we would have to consider the peak-to-peak noise in that case.

    It seems like you will have to move to one of our higher-performance ADCs in order to achieve the performance you are looking for. The ADS124S06 should be able to meet your targets when operating at the lower data rates.

    Regards,

  • 0 in reply to Joachim Wuerker
    Prodigy 40 points
    Hi Joachim,

    Thanks for your reply, you guessed it right what I want is 1mK resolution for that range of temperatures. I don't necessarily need them to be noise-free since I was thinking in filtering the signals into the MCU anyway.

    I'm checking the ADC you suggested and from the "Resolution from VRMS" tables I see is far superior. Would I be able to do a ratiometric measurement of two 2-wire PT100 as in the document I linked? If it's not too much to ask, could you provide an optimal configuration as you did before to take as a reference?

    And I have a side question, when I was checking the ADS1220 datahseet I calculated the common mode requierements and I got these numbers for AVDD=3.3V AVSS=0V PGA GAIN=16 Iex=1000mA.

    VCM min condition1 > 1.14V
    VCM min condition2 > 0.825V
    VCM max < 2.15V

    VCM I get is around 0.17V since the PT100 produces such a tiny signal, so, I guess that enabling the PGA to measure a PT100 is out of question?

    Thanks for your time and best regards
  • 0 in reply to Pablo G.C
    Prodigy 40 points
    I meant Iex=1000uA , not mA
  • 0 in reply to Pablo G.C
    TI__Mastermind 23495 points

    Hi Pablo,

    yes, you could in general implement the same circuit with ADS124S06 as shown in the TechNote as well.
    There are two minor differences though:

    • With ADS124S06 you could only use a gain = 1 with this implementation, because the device only offers gain = 1 when bypassing the PGA.
    • The IDAC (excitation current) in ADS124S06 cannot be routed internally to REFP0, means you need to use another analog input (e.g. AIN4), tie it to REFP0 and route the IDAC internally to AIN4.


    The only reason why we implemented the circuit as shown in the TechNote with the two RTDs in series is, that this is the only way we can measure two 2-wire RTDs with ADS1220 without having to use an external multiplexer.
    This circuit is targeted for Heat Meter applications where cost pressure is extremely high.
    The solution as shown in the TechNote requires that the PGA inside ADS1220 be bypassed. As you realized, we wouldn't meet the common-mode voltage requirement of the PGA if we were to use it. Therefore we are limited to gains of 1, 2 and 4 in ADS1220.
    We could level-shift the whole circuit by placing a bias resistor between the lower RTD and GND. However one other challenge in Heat Meters is that they are battery powered and often operate from a supply voltage of 3V or less. Therefore it gets challenging to make the circuit work with the additional voltage drop across the bias resistor.

    With ADS124S06 you are much more flexible in the way you can measure two RTDs because the device offers more analog inputs.
    The most common approach would be to have both RTDs more or less in parallel and place the reference resistor on the low side.
    Following is a principle drawing of this implementation.


    The reference voltage created across the reference resistor acts as a level shift for the RTD measurement. This allows to use the PGA and higher gains.
    The following settings would seem to be a good starting point given your requirements (Pt100, T_RTD = 0°C to 45°C, AVDD = 3.3V):

    • IDAC = 1.5mA
    • Gain = 8
    • RREF = 1kOhm

    According to my calculations that should yield a temperature resolution of about 0.2mK_rms when using a data rate of 10SPS. If you use a lower data rate it will improve more.
    Configurations with smaller excitation currents will yield marginally worse resolution. We probably don't want to increase the excitation current much beyond 1.5mA to limit the effect of RTD self heating.

    It is important to note that at any given time only one RTD is to be excited with the IDAC. When you direct the IDAC to the 2nd RTD, you need to make sure the signals settle before starting the conversions.

    You could try the circuit out on our ADS124S08EVM to see if you are getting the performance you need.

    Regards,

  • 0 in reply to Joachim Wuerker
    Prodigy 40 points
    Hi Joachim, that looks good enough to me and that evaluation board really comes in handy, thanks and best regards.