• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 46 subscribers
  • Views 944 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.

need help for AFE design using ADS1263

Xu Junhua
Prodigy 90 points
Other Parts Discussed in Thread: ADS1263

Hi，

I‘m developing one DAQ system based on ADS1263, which needs one precision and low noise AFE.

The AFE could condition the input voltage signal from 1mv to 300v.

Could anyone provide the reference design?

Thanks，

Junhua

  • 0
    TI__Mastermind 44680 points

    Hi Junhua,

    Could you describe the input signal with more details?

    - Are you trying to measure 1mV to 300V differentially, or is this the common-mode voltage range?

    It would be more common to handle this wide of a common-mode voltage, as opposed to a differential voltage. However, if this is the differential voltage, then it would need to be significantly attenuated! 

     

    - Is the signal single-ended, pseud-differential, or a true differential signal? In other-words, what is the input voltage referenced to?

     

    - Where is this voltage coming from and what is the sources's output impedance?

     

    - Is the signal a DC voltage or are there AC components in the signal as well? How much bandwidth is needed to capture the signal? 

     

    - Will you need to use isolation (on the analog or digital side)?

     

    - Will you have multiple input channels (do they need to be isolated from each other)?

     

    Best Regards,
    Chris

  • 0 in reply to Christopher Hall
    Prodigy 90 points

    Hi, Chris,

    please read the comments.

    Could you describe the input signal with more details

    Actually, we are designing one 6.5 bit DMM, and the input signal should been conditioned to +-2.5V, and single-ended input, due to the AD is powered by single +5V.

    - Are you trying to measure 1mV to 300V differentially, or is this the common-mode voltage range?

    It would be more common to handle this wide of a common-mode voltage, as opposed to a differential voltage. However, if this is the differential voltage, then it would need to be significantly attenuated! 

    the input signal could be  differential signal, and the voltage scope is: -100mV~100mV, -1V~1V, -10V~10V, -100V~100V and -300V-300V.

    - Is the signal single-ended, pseud-differential, or a true differential signal? In other-words, what is the input voltage referenced to?

     the input signal is connected to the AD by single-ended mode, which is shifted up to 2.5V by the AD's reference voltage.

    - Where is this voltage coming from and what is the sources's output impedance?

     the source's output impedance is unknown, but the AFE's input impedance should be bigger than 1M.

    - Is the signal a DC voltage or are there AC components in the signal as well? How much bandwidth is needed to capture the signal? 

     the input signal could be DC and AC, but he bandwidth is <1K.

     

    - Will you need to use isolation (on the analog or digital side)?

     no isolation , but the over load protection is needed.

    - Will you have multiple input channels (do they need to be isolated from each other)?

    only one channel.

    Thanks,

    Junhua

  • 0 in reply to Christopher Hall
    Prodigy 90 points

    Any update for my comments in line?

    Thanks,

    Junhua

  • 0 in reply to Xu Junhua
    TI__Mastermind 44680 points

    Hi Junhua,

    Thanks for your patience! I was looking through the TI Design library to see if we had any relevant reference designs that could be helpful to your application but didn't find anything that showed how to measure voltages up to +/- 300V or protect against over voltages that large. Therefore, please allow me to provide my thoughts on the AFE design...

     

    I started by looking into a few options for attenuators and after comparing some passive and active options, I think the best options are either to use a 1) passive voltage divider, or possibly a 2) difference amplifier to attenuate the high voltage signals down to the +/- 2.5V or +/- 5V range (Note: An external 5V-reference would be required to measure a +/-5V signal; however, there is a good chance you would need a higher precision external reference source in a DMM application anyways). The difference amplifier circuit would require twice as many resistors; however, these resistors are smaller values and would have lower thermal noise. On the other hand, the passive circuit doesn't have the added offset and noise coming from the amplifier. Therefore, the op-amp selection would be an important deciding factor between these two circuits:

    From there, I would likely design each input range setting circuit and provide a low-leakage, low thermal EMF relay to switch between the input circuits (or switch in the correct resistor values for the desired attenuation).

    By switching between inputs circuits, each circuit could have its own input protection circuitry designed to clamp voltages outside of the input range. Alternatively, you might consider trying to design a high voltage current-limiter that would limit the input current to some reasonable value. An example of a current limiter (albeit for a lower voltage application) can be found in the following TI Design:  

    I hope that helps! Let me know if you have any follow-up questions and I'll try my best to answer them.

    Best Regards,
    Chris

  • 0 in reply to Christopher Hall
    Prodigy 90 points

    thanks for your help, your idea is good.

    BTW, do you have any solution about measurement of electrical resistance, that should be from 0~100M。

    Thanks,

    Junhua

  • 0 in reply to Xu Junhua
    TI__Mastermind 44680 points

    Hi Junhua,

    The most common way I know of would be to force a current through the unknown resistance and measure back the voltage. The current amplitude would need to be small (and perhaps limited) to avoid generating a large voltage, so this method may not be the best at measuring large resistances. However, if you're able to provide a feedback loop to increase the current for smaller resistor values, then this method would have better sensitivity for measuring small resistor values.

    Alternatively, you could apply a voltage and measure back a current; however if the resistance is small, this can result in generating a very large current. Therefore, this current would need to be limited and it would be difficult to measure smaller resistances using this method. However, a second precision resistor (in series with the unknown resistor) would be able to serve the purposes of both limiting and measuring the current flowing through the unknown resistor. By increasing the maximum allowable current, you will increase the measurement sensitivity for smaller resistor values.

    Some resistance measurement instrumentation will allow for both of these types of measurement techniques. While still others types of instrumentation may connect the unknown resistance to a bridge and attempt to balance the bridge to determine the value of the resistor.

    Here are some references that might be helpful:

    • Different Implementations of current sources:  
    • Keithley (now Tektronic) application note on High Resistance measurements:  
    • Agilent Impedance Measurement Handbook:  

    Best Regards,
    Chris