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.

ADS1248: how many 4-wire RTDs measurement possible with ADS1248?

Part Number: ADS1248
Other Parts Discussed in Thread: ADS124S08

This is a simplified diagram of the similar configuration shown in the main question but with two RTD in series, each pair used with one IDAC and REF inputs. A similar configuration is shown in TechNotes SBAA235.

My question here is if this is possible? if yes, then please comment on the accuracy of such implementation considering we provide the recommended filtering and drive the ADC at sub 30 SPS speed. I am trying to get < ±0.6°C accuracy with the maximum number of RTD possible. 

Regards,

Yash RK

  • Yash,


    I think this type of setup should get you the accuracy that you're looking for. I'll need more information after you've decided the RTD type and accuracy, the data rate, and the reference resistance that you'll select. However I think ±0.6°C accuracy is very achievable. Again, you'll still need to use RTDs of a higher tolerance class so that the RTD accuracy itself isn't in question.

    I would note that you're limited by the number of inputs that are available, the ADS124S08 is available. This device is considered the next generation of the ADS1248. If you're interested, go to the product folder and look at the datasheet. Based on a quick look, you should be able to get an extra RTD measurement (basically five 4-wire RTD measurements instead of four).


    Joseph Wu
  • Joseph Wu thank you for your reply.

    From what I can see from the datasheet, to have two reference input we need to use AIN6 & AIN7 and then two current sources will be needed so to get that we will again need to use to analogue pins (say AIN8&9) which again makes the maximum number of RTD to four.

    Or am I missing something?
  • Yash,


    Sorry about that, you are correct. I had drawn it out quickly. I started with AINCOM for one IDAC source and combined AIN0/1, AIN2/3, and AIN4/5 to use with AIN6/7 as the reference input. I then combined AIN8/9 and AIN10/11 for REF0, but forgot I needed an extra pin to source current for the second IDAC.


    Joseph Wu
  • Dear Joseph,

    I have selected the value of reference resistor and want to get your opinion on it.
    I am using a bipolar +/-2.5V supply as I will be also using inputs of the same IC as single ended inputs apart from the above mentioned RTD design and I am switching between them.
    I am planning to use 735 ohms resistor as reference resistor with 1mA current flowing through it with 4v/v gain in PGA.
    Considering my application maximum temperature swing will from 0-200 Deg Celcius so RTDmax = ~180ohms.
  • Yash,


    The values that you mention should work, but it would be nice to use a larger reference. Because you have the bottom of the reference tied to analog ground and the supplies are bipolar, it means that you have the reference and RTD voltages within the range of 0V to +1.5V (removing 1V from the positive supply because of the IDAC compliance). PGA range is ok, the reference resistor and gain should maximize the resolution of the 180 Ohm RTD resistance.

    That setup is near the minimum reference voltage of 0.5V, but again it, should work. However, there are two alternatives to make the reference larger and fit within the operation of the ADC. First you could use a unipolar supply. This would allow more room for both RTDs and the reference resistance. Alternatively, you could tie the negative reference to AVSS. This would accomplish the same.

    Regardless, it would be nice to test this operation. If you want, you could get an ADS1248 EVM to make a few measurements. Note that the software for the evaluation module may not run on Windows 10 (it should be ok for Win 7). When you get a more complete schematic, it would be good to review it.


    Joseph Wu
  • Thank you for your reply, Joseph.

    1. The first option you mentioned was to use unipolar supply, in that case, how can I provide 2.5V on the AINn (without using up another analog pin) for my unipolar signals.

    2. For your second option, even if we tie the Vss to -2.5V, yes we will get a more headroom as max we can go to is 1.5V but increasing the reference resistor (in either case) will still require us to increase gain in PGA right? (As I was trying to get away with lower gain as possible)

    And the solutions you gave are just to make sure we stay in compliance voltage and not to increase the resistance without increasing the gain if I understand you correctly. Also, I would like to send you the schematic in private to show you the whole implementation, but I don't know how can I send it?

    Regards,

    Yash.

  • Yash,


    1. I'm not sure I understand this point. Why do you need to provide 2.5V to the AINN? Everything is already biased with the IDAC currents and the resistors.

    2. It is a valid point that if you increase the reference, you'll also need to increase the gain which will increase the noise based on the resolution. For the resolution, I mean the noise compared to the full scale range. However, if you look at the noise tables in the datasheet, there really isn't much difference in the noise for PGA gains 1 to 8. This way, you can increase the gain, and reduce the effect of any noise that you might get from having a small reference.

    Regardless, I've sent you a message through E2E messaging. We can discuss this further there and you'll be able to send a schematic without making it public.


    Joseph Wu