• State Resolved
  • Locked Locked
  • Replies 13 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 1040 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.

ADC12D1620QML-SP: Absolute Maximum Ratings

Richard Raffanti
Prodigy 160 points
Part Number: ADC12D1620QML-SP
Other Parts Discussed in Thread: LMH5401, LMH5401-SP

Hello, 

I'm trying to understand the absolute maximum ratings for the analog inputs of this device.

In Table 6.1 of the datasheet, the abs max for the four Vin pins is given as -0.15v to +2.5v with the note "while maintaining common mode" and the additional note 3 saying that this is verified at VA = VD = 2.2v.  Normally I would expect that an absolute maximum rating expressed this way (as an absolute value, 2.5v, not related to supply rail) would be valid at any permissible supply voltage, but note 3 makes me wonder (with a 2.2v supply, we can expect the protection diode will not be conducting much for Vin = 2.5v).  That's my first question: is the +2.5v absolute maximum for the four Vin pins valid irrespective  of the value of VD or VA?

My second question relates to the -0.15v rating.  In Table 6.3, Recommended Operating Conditions, the minimum voltage at the four Vin pins is specified as -0.4v.  This seems to violate the abs max rating on the face of it, so I wonder how to reconcile that.  That's my second question: Is the -0.15v absolute maximum rating valid? How can the recommended be outside the absolute max?

I am designing a data converter system for a Mars exploration mission, and I'm needing to prove device safety for anomalous conditions, such as during power up or test, when one supply of the driving device may be unpowered.  In this case both inputs might be driven above the 1.9v rail or below ground, but there are series resistors which would limit the current to below the +-50mA maximum, assuming the protection diodes turn on.  But I don't think it's possible to limit that voltage to -0.15v without some other clamp.  And, if the +2.5v rating should really be expressed as VA + 0.3v, the same problem applies to the positive excursion.

Thanks for any help

Rick Raffanti

Techne Instruments, Inc

(contractor for Southwest Research, San Antonio)

  • 0
    TI__Mastermind 43930 points
    Hi Rick
    I'm reviewing your questions and working on a response.
    For your application do you need to DC-couple to the ADC inputs, or would AC-coupled inputs be acceptable? AC-coupling can remove many of the DC stress that need to be considered.
    Best regards,
    Jim B
  • 0 in reply to Jim Brinkhurst1
    Prodigy 160 points

    Hi Jim,

    I thought I posted a reply to your question a couple days ago, but I don't see it here.  In any case, my application is DC-coupled, using the LMH5401 as a driver.  We use the VOCM from the ADC to establish the common-mode voltage.  We are looking at unipolar signals from a PMT sensor in a mass spectrometer, and we want to use as much of the ADC dynamic range as possible, even if our duty ratio is varying.  Thus we need to DC-couple

    THanks for your help.

    Rick

  • 0 in reply to Richard Raffanti
    TI__Mastermind 43930 points

    Hi Rick

    Thanks for the additional information. That will help with my response.

    I want to point out an important detail when using the LMH5401 to drive the ADC12D1620 and ADCs with similar common mode voltages.

    To get best performance from the LMH5401, the power supply rails of the amplifier should be centered around the output common mode voltage. If the output common mode is significantly offset from the supply voltage center point the harmonic distortion performance of the amplifier will be degraded. There are several datasheet Typical Characteristics plots showing Harmonic Distortion Performance vs Output Common Mode Voltage which help illustrate this effect.

    Best regards,

    Jim B

  • 0 in reply to Jim Brinkhurst1
    Prodigy 160 points
    Hi Jim,
    Right, we are planning to use +3.75, -1.25 with Vcm established by the ADC. Thanks for calling my attention to those plots in the 5401 datasheet. We are planning to adjust out the typical 27mv offset from the 5401 Vicm to Vocm, though now that I've noted the Vcm+-150mv recommended operating condition of the 1620, I suspect that's not necessary.
    Thanks,
    Rick
  • 0 in reply to Richard Raffanti
    Prodigy 160 points

    Hi Jim,

    Any further thoughts on this?

    Thanks

    Rick

  • 0 in reply to Richard Raffanti
    TI__Mastermind 43930 points
    Hi Rick
    Sorry for the delay. I've been delayed somewhat due to business travel, but I am still working on the question.
    I do have some concern about a case where only one of the amplifier supplies is active.
    Would it be possible to gate the supply voltages to the LMH5401 so they are only enabled when the ADC 1.9V supply is active and both the +3.75V and -1.25V supply sources are regulating?
    Best regards,
    Jim B
  • 0 in reply to Jim Brinkhurst1
    Prodigy 160 points

    Hi Jim,

    It would be good to sequence the supplies in such a way that the front-end supplies cannot be powered up unless the ADC supplies is up, and maybe we can do that (although the two boxes are separated by a couple meters of cable).  But I think the concern is for the anomalous situation, perhaps in test, when something unusual happens.  In that case, we want to prevent damaging flight parts.

    I'm hoping that the series resistance of our input attenuator will be enough to limit the current through the ADC clamp diodes to less than the +-50mA maximum.  Here's our system design

    Rick

  • 0 in reply to Richard Raffanti
    Prodigy 160 points

    There's a little typo in that drawing- that middle resistor in the T attenuator should be 140 ohms, not 16.5 (it's a 6dB differential attenuator).

  • 0 in reply to Richard Raffanti
    TI__Mastermind 43930 points
    Hi Rick
    Thanks for the additional detail.
    I'll work through it and get back to you tomorrow.
    Best regards,
    Jim B
  • 0 in reply to Jim Brinkhurst1
    TI__Mastermind 43930 points
    Hi Rick
    Sorry for the delay.
    I've reviewed your schematic details. Based on my understanding of the LMH5401-SP output circuitry I believe the ADC inputs will not be driven beyond the allowed voltage or current conditions even if the ADC power is off, and the amplifier is fully or partly (only 1 of the 2 supplies are active) powered. This analysis would not cover cases where the amplifier supply voltages are significantly higher than +3.75V or lower than -1.25V.
    Best regards,
    Jim B
  • 0 in reply to Jim Brinkhurst1
    Prodigy 160 points

    Hi Jim,

    I'm still troubled by the Vss-0.15v absolute maximum on the ADC inputs.

    A colleague of yours (through another channel) suggested a different approach, which also saves us some complexity: use +5, 0v supply for the LMH5401, set VCM to +2.5v, then use a 6dB attenuator in each leg to reduce VCM to the ADC to 1.25v.  I've simulated this and it seems like a better approach to us.  This will result in a max voltage applied to either input of 2.5v, which won't exceed the absolute max rating (maybe we'll make it 7dB to get a little margin).  This assumes that +2.5v abs max spec can be taken as written, not dependent on VA or VD.  Can you confirm that?

    Thanks,

    Rick

  • 0 in reply to Richard Raffanti
    TI__Mastermind 43930 points
    Hi Rick
    That seems like a reasonable alternative approach, and simplifying the power supplies is also a good reason to use it.
    Per the cold sparing section of the datasheet, it would be best to maintain the maximum voltage at the VIN+/- pins to 2.15V for any potentially continuous/longer term events. What attenuation would be required to achieve that?
    I will discuss this with my LMH5401-SP colleagues as well.
    Best regards,
    Jim B
  • 0 in reply to Jim Brinkhurst1
    Prodigy 160 points

    Thanks Jim.  I see now that the 6dB will be safe, since the LMH5401 can only swing to +3.9v, and therefore the ADC input will only see +1.95v max.  That's what we'll do.

    Best regards,

    Rick