• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 45 subscribers
  • Views 154 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

ADS114S08: design help for reading poteniometer(s)

Jeevan Ghimire
Prodigy 10 points
Part Number: ADS114S08
Other Parts Discussed in Thread: TPS22946

Hi here is a quick overview of the system.

I am designing a battery powered prototype of potentiometer(s) readout. The device displays the potentiometer wipers ratio value in an LCD and goes to sleep after a timeout to conserve battery life.

The potentiometers are around 50-100m away from the ADC input pins. There can be anywhere between 1-10 pots connected to ADC. They are 10K pots. I have provisioned 2 inputs as lead wire between excite and ground terminal of pots to be measured as differential inputs. The 10 channels are read as single ended.

Pots are to be excited by 3.3V with the same 3.3V also used as reference inputs REF0 for ratiometric measurements.

1) SPI communication: is the schematic correct (as per the datasheet) or do we also need to connect RESET pin to 3.3V with 10K pot? I want microcontroller to be able to reset the adc in case of any issues.

2) Short: I want to provide some short circuit protection and detection between pot excitation and GND. Since the effective resistance between the excitation and GND can be anywhere between 1KOhms (with 10 pots connected) to 10KOhms(with 1pot connected), I think the max current draw should be between 3.3mA - .33mA.

I was thinking to implement a switched excitation source(3.3V_Excitation) where first we check for short circuit and only then provide the excitation for the pots. This is to protect the circuit and optimize the battery life by protecting against the unwanted current draw.

I found TPS22946 as the smallest switch with fault detect and shutdown option(for no pots connected and device in sleep). I was thinking of using this to protect against and detect the short (i.e. when the current exceeds 30mA limit). However my maximum current draw should not exceed 3.3mA so anything more than that is error condition. Is his correct? Are there any similar devices/switches that you can recommend which fits my requirement? Is this the correct approach.
Also I found out the burntout detect feature of this ADC but the datasheet says its difficult to detect cable shorts using this feature. I also did not find much info on how I can implement this feature especially in my case.

3) The wipers do not move too quick. I can get away with 5-20 SPS and display the average of the readings to be displayed in LCD. Which sampling should I use to get the best noise rejection? Any tips for input protection in my case. I was thinking of protecting all the 14 I/O pins to the pots with TVS diodes (3.3V /5V). Any suggestions for a standard protection system?

  • 0
    TI__Guru** 112745 points

    Hi Jeevan,

    Welcome to the E2E forum!  What you are proposing is very similar to a 2-wire RTD or thermistor measurement circuit that is voltage excited.  I would recommend the following pieces of information:

    I think in the end you will have lots to overcome due to the cable length including cable impedance, RFI/EMI and power line-cycle noise. As the wiper approaches AGND, it will become more and more difficult to overcome the noise.

    Here are my additional responses to your questions:

    "1) SPI communication: is the schematic correct (as per the datasheet) or do we also need to connect RESET pin to 3.3V with 10K pot? I want microcontroller to be able to reset the adc in case of any issues. [Bob] I would recommend pulling the RESET pin high with a weak pullup resistor (10k to 100k) and use the micro to pull the pin low if needed.

    2) Short: I want to provide some short circuit protection and detection between pot excitation and GND. Since the effective resistance between the excitation and GND can be anywhere between 1KOhms (with 10 pots connected) to 10KOhms(with 1pot connected), I think the max current draw should be between 3.3mA - .33mA.

    I was thinking to implement a switched excitation source(3.3V_Excitation) where first we check for short circuit and only then provide the excitation for the pots. This is to protect the circuit and optimize the battery life by protecting against the unwanted current draw.

    I found TPS22946 as the smallest switch with fault detect and shutdown option(for no pots connected and device in sleep). I was thinking of using this to protect against and detect the short (i.e. when the current exceeds 30mA limit). However my maximum current draw should not exceed 3.3mA so anything more than that is error condition. Is his correct? Are there any similar devices/switches that you can recommend which fits my requirement? Is this the correct approach.
    Also I found out the burntout detect feature of this ADC but the datasheet says its difficult to detect cable shorts using this feature. I also did not find much info on how I can implement this feature especially in my case. [Bob] Review the material I provided earlier to see if one of the methods described will work for you.

    3) The wipers do not move too quick. I can get away with 5-20 SPS and display the average of the readings to be displayed in LCD. Which sampling should I use to get the best noise rejection? Any tips for input protection in my case. I was thinking of protecting all the 14 I/O pins to the pots with TVS diodes (3.3V /5V). Any suggestions for a standard protection system? [Bob] The lower the data rate the lower the input referred noise.  See the noise tables in the device datasheet.  It is good to use an RC filter to act as an antialiasing filter and also as a current limit to the ADC analog inputs.  TVS diodes are a good idea, however you must also make sure that the Absolute Maximum Ratings of the ADC are not exceeded.  In addition to TVS diodes you may need to add additional filtering such as ferrites.  The design for input protection should be based on the conditions expected for the system operation.  You probably need to have shielded cabling and the shield should be properly terminated."

    Best regards,

    Bob B