• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 46 subscribers
  • Views 512 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.

Original question:

ADS131E08: are there any clamp diodes at the analog input pins?

ADS131E08: ADS131E08 Dying inputs random

Keith Wakeham
Prodigy 10 points
Part Number: ADS131E08
Other Parts Discussed in Thread: ADS1298

I've been using the ADS131E08 on several strain gages with an external reference pulled from the bridge (6 wire strain gage config, power/gnd,2 sense, 2 analog -- well many analog). The power supply is a simple unipolar with AVSS to GND and AVDD to 3.6-4.2v (lithium battery) with some isolation. It has worked very well for accuracy but now after months of usage of the prototypes the client has experienced random death of random inputs. The reference is being pulled from the strain gages and is about 0.1V less than the AVDD to GND so it's within the allowable. We actually have two, the primary generating the clock for the secondary.

I omitted the external clamping diodes thinking they shouldn't be needed and then thought that maybe there was no ESD on the analog inputs at all but several forum posts indicate there is. Is this chip not robust?

At first a channel died on the second chip on a board, then another had 4 dead over two chips. These worked for months up until this point but now most are showing signs of this death and one recently started to overheat. We've been unable to replicate this here with our boards at all.

We only have a very basic filter on the strain gages, and the analog rail is controlled by a P-Fet high side switch. I believe I replicated the unipolar reference faithfully, but have I done something dumb here? Is there any way this could be ESD?

Note the schematic says ADS1298 but these are ADS131E08. Sloppy not updating on my part.

  • 0
    TI__Guru 66346 points

    Hello,

    Thank you for your post and welcome to our forum.

    The ADS131E08 has limited inherent ESD protection. The HBM and CDM ratings are provided in section 7.2 of the data sheet. It is still entirely possible that an external ESD event could damage the analog inputs without an external ESD protection circuit.

    You mentioned some devices are overheating - are you noticing an increase in supply current to the device? Are the affected channels able to recover after resetting the ADC or power cycling the unit, or does it appear permanently damaged?

    Regards,

    Ryan

  • 0 in reply to Ryan Andrews
    Prodigy 10 points

    Hi Ryan,

    There is a notable increase in current. They did indicate that they were charging the battery while using the device and the the overheating events only happened during these periods. While I wasn't keen on this operation, the charger shouldn't be able to put out more than 4.2v. I suspected that maybe an improperly isolated AC to USB adapter could have caused the faults.

    The Analog inputs haven't recovered when fully disconnected and reconnection of power. The registers write and then readback so I don't think it's faulting there. But once we've lost a channel it tends to not come back. With a gain of 12 (max for the adc) it is often near railed but not quite. They sort of read signal, but not really.

    As for the overheating / high current draw, that seems to go away on power disconnect and reboot. It was suggested that maybe the damage could be causing SCR latch up.

    -Keith

  • 0 in reply to Keith Wakeham
    TI__Guru 66346 points

    Hi Keith,

    It sounds like a latch-up condition since the device is not recovering after power cycling. The interface and register map seem to be functional. Some clamping diodes or current limiting on the analog supply may be needed to ensure the Absolute Maximum Ratings are not violated while charging the battery.

    Are you able to monitor the voltages on the AVDD and AVSS pins while charging?

    Regards,

    Ryan

  • 0 in reply to Ryan Andrews
    Prodigy 10 points

    Finally got a broken board from the customer. We're seeing the input pins having a strong short to AVSS via pins 32, 57, 58,  in the 170ohm range while were seeing around 170kohm to AVDD from these pins. The good pins are showing >5M. The other AVSS pins 21 and 23 show a 170-190kohm short with these pins. So not all AVSS are equal. It's beginning to sound like some sort of overvoltage or ESD damage. Any suggestions on checking or replicating?

  • 0 in reply to Keith Wakeham
    Prodigy 10 points

    And now I tried 7 - 12 on the inputs, just glancing on this as it's sacrificial hardware. Was able to cause 170kohm then 170ohm shorts to AVSS..... I think this confirms ESD damage from handling

  • 0 in reply to Keith Wakeham
    TI__Guru 66346 points

    Hi Keith,

    The low impedance measurements from the inputs to AVDD and AVSS also indicate to me that the devices are being damaged. This can be caused by an overvoltage or overcurrent event, such as ESD. If this continues to be a problem, the customer may need to modify their design to include additional ESD protection diodes.

    Regards,

    Ryan