Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

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.

ADS1256 Input protection

Other Parts Discussed in Thread: ADS1256

Dear all,

i am using an ADS1256 in my project.

Last days one ADS was destroyed with following malfunction:

The reading of the 2 used channels was only approx. 75 % - 85 % of what it should be.

Then I have measured a voltage drop from 1.25 V down to 1.1 V over the input protection series resistors of 100 Ohm on both channels.

So the input current into one channel is approx.. I= 0,15V / 100 Ohm = 1.5 mA !

The analogue inputs seems to have had a low resistance.

I switched the whole unit off and on again, but still have the same behavior. So it was not a latch-up ?

 

After unsolder and putting a new ADS1256 everything works fine again without any software change.

Now I am thinking about the reason of what has destroyed the ADS chip.

If I look at the maximum ratings of the ADS1256, the maximum voltage to the analogue inputs is AVdd + 0,3 V.

I am using input protection diodes type BAV199 to +5V and GND, which have a low leakage current to get the best performance of the system.

As these type of diodes are not Schottky, they have a forward voltage drop of 0.6…0.8 V according the data sheet.

So in case of a short circuit of the external sensor, the amplifier will supply +12 V to my ADS1256 input protection circuit. As a result, the voltage on the analogue input pin can be 5.7 V (not measured). Current into the ADS pin is limited to less than 10 mA . I = 0,7V / 100 Ohm= 7 mA.

Schottkey diodes have a lower forward voltage drop to give a better protection, but the leakage current is much more higher.

So my question is, can my poor protection circuit be the reason for damaging the ADS ?

What is the perfect way for good and accurate input protection of the ADS1256 ?

  • Hi Dirk,

    Welcome to the TI E2E forums and sorry for the delay!

    Which resistor did you measure the 1.25V voltage drop across?
    For instance, if you measured 1.25V across R32, then you'd have 1.25V / 100Ohms = 12.5mA of current going into the ADC inputs. This could potentially damage the device. However, if this was measured across R29, then we can only try to estimate how much current is sinked by the ADC and how much is sinked by the protection diode...

    Here is my estimation...
    When the input voltage exceeds the supply voltage, the internal ESD diodes inside the ADS1256 will begin to turn on and the inputs will become low impedance. Assuming that the BAV199 has about a 1V forward voltage drop, you would then have about 6V present at the ADC input. Assuming that the internal ESD diode is a low impedance connection to AVDD (5V), there is only about 100 Ohms of impedance between the 6V input and 5V supply rail, or about 10mA of current. This puts you right on the edge of exceeding the absolute maximum ratings, and I've only looked at the static behavior of this circuit. The transient behavior could be worse if the BAV199 protection diode doesn't turn on as fast as the ADS1256's internal ESD diode; therefore, the ADC input could initially see more than 6V at the input.


    I would recommend increasing the R32 and R35 resistors a bit.
    At 200 Ohms, the steady state behavior would limit current to 1V / 200Ohms = 5mA. However, if you can increase these resistors to 1.2kOhm, then it would guarantee that the ADS1256 does not see more than 10mA of current in the case that the BAV199 is slow to turn on. I wouldn't recommend going much higher than a few kOhms, since resistor thermal noise starts to approach the ADC's input referred noise performance usually around 3kOhms to 5kOhms.

    Best Regards,
    Chris

  • Dear Chris,
    thanks for your good support.

    >>>Which resistor did you measure the 1.25V voltage drop across?
    Yes it was resistor R32 and R35. Of course your calculation is correct (12.5 mA).

    I have followed your suggestion and have already changed R32 and R35 from 100 Ohm to 240 Ohm on my prototype board.
    Well I still have the noise free resolution of 0.5 PPM. That is fine.

    Maybe other reason for the damage was, that I have 2 separate ground planes (digital and analogue). The are not connected together on the PCB, only inside the power supply. So I have made additional a spldered jumper for both ground planes direct under the ASD. Nothing has changed in the good noise free resolution.

    So i hope with these 2 changes my baord will run without problem in the future.
    Thanks again.

    Kind greetings from Germany. Dirk
  • Hi Dirk,

    Yes, keeping a short low impedance connection between analog and digital ground should also help.

    Please let me know if I can be of any additional assistance!

    Best Regards,
    Chris