• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 47 subscribers
  • Views 242 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.

LM747: Discrepancy in observed transient response overshoot between two lots of LM747H chips

Tyler Garlock
Prodigy 10 points
Part Number: LM747

The datasheet for the LM747 dual-opamp lists only typical values for Transient Response: Rise Time (0.3 µS)  and Overshoot (5%).  Is there a reason why no maximum values are given here?  We have 2 lots of TI (then National Semiconductor) LM747H chips.  One lot, manufactured in 1984, has observed rise times of 0.16 - 0.20 µS with overshoots of 3.5-6%.  The other lot, manufactured in 1994, is showing similar rise times, but has observed overshoots of 18-20.5% for all units with similar rise times as the other lot.

Is there a reason why such a discrepancy in overshoots would exist between two of (what should be) the same lots of parts?  And since the 5% listed in the datasheet is only a typical value, is a ~20% overshoot actually an issue with the part itself?  Unfortunately, I don't know the end-use application, so I can only speak in terms of the chip itself, now how it will act in its final circuit. 

Thank you,

Tyler

  • 0
    TI__Guru**** 151036 points

    Hello Tyler,

    The data sheet specification is unity gain and no capacitive load. For any other setup, the result could be different.

    For no capacitance and a 2k resistive load, this chart could have a a greater phase shift or an extended unity gain frequency. Either or both would  increase overshoot.

  • 0 in reply to Ron Michallick
    Prodigy 10 points

    Ron,

    Thank you for your response. Based on what you said, a circuit 1) with no capacitive or resistive load and 2) that operates the Op-Amp with a unity gain should yield "typical" results. The circuit I'm using meets both of those criteria. For one lot of amplifiers, I'm getting expected results, but I'm still not sure why one of my amplifier lots (6 units with date code H94AB which I assume means manufactured in '94) are seeing a 20% overshoot, while the others manufactured, manufactured in '86, are staying around 5%. 

    Is this kind of discrepancy between two date codes, using the same test setup, generally acceptable?

    Thank you,

    Tyler

  • +1 in reply to Tyler Garlock
    TI__Guru**** 151036 points

    Tyler,

    While the difference is significant, the data sheet defines what is acceptable or not. There is no limit unless the part number is LM747A/LM747E. That limit is 20% 

  • 0 in reply to Ron Michallick
    Prodigy 10 points

    Ron,

    That answers my main question, but as a bonus: If we had an application that required the 20% max overshoot limit, we would have to order the LM747A, because chips with that suffix are actually tested to meet that spec?  Thank you.

  • +1 in reply to Tyler Garlock
    TI__Guru**** 151036 points

    Tyler,

    That is a good question but I don't have an answer. I can't find build instructions for the "A" version.

    I do not see a public option to buy LM747 or LM747A from Texas instruments.