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.

TLV9062: variance of the common mode voltage Vs Offset voltage

Part Number: TLV9062
Other Parts Discussed in Thread: OPA365, OPA364, OPA325, OPA388, OPA320, OPA322

Hello,

 

There is a graph showing the variance of the common mode voltage Vs Offset voltage (see below copied from the datasheet). This is of concern to us as if we have a signal feeding into this amp as a unity gain buffer (for example) if that signal rises from 3V to 4.5V then according to this there will be a point where the offset voltage will step (if I am reading this right)

 

What are the different coloured lines for? I have assumed temperature but there is no key and I would be interested in which temperatures map to which lines.

 

Further I want to confirm I am reading this correctly, this means that there is a step in offset voltage when running as a unity gain buffer at ~4.1V – is that correct? (assuming V- = 0V and V+ = 5V). So in the situation where a signal is varying around this level there would be an extra 1.5mV of noise added to the signal?

 

 

  • Sorry the file didn't seem to attach properly in the original message. see below for the datasheet graph

  • Hi Ben,

    you the see the performance of several typical TLV9062. Each coloured curve belongs to such a TLV9062.

    Yes, there's a step in input offset voltage, when the comon mode voltage reaches a certain level. Section 9.3.1 of datasheet explains why this happens.

    Kai

  • Hello,

    So to be clar the differnt lines in teh graph are showing manufacturing variances?

    Kind regards,

    Ben Strawbridge

  • Ben,

    Kai is correct - Thanks Kai!

    This behavior, called input crossover distortion, is common to most general purpose and precision rail-to-rail input products. If this behavior is undesirable consider using a "Zero-Crossover" op amp such as the OPA320, OPA322, OPA325, OPA365, OPA364, or OPA388. These devices feature an internal charge-pump that eliminates the need for the complimentary input stages. The OPA325 has the same 10MHz GBW as the TLV9062.

    Here's a tech note that describes zero-crossover op amps.
    www.ti.com/.../sboa181a.pdf
  • Hello Collin,

    Is there a figure for the worst case Vos? the datasheet says ±2mV across the temperature range, does this include the variance at the cross over point? I assume not as it states this is at Vs/2. so is there a worst case figure for this available?

    I am just trying to establish the worst case.

    Kind regards,

    Ben Strawbridge
  • Hi Ben,

    your observation is correct. The input offset voltage is only specified for Vs/2.

    Keep in mind that the TLV9062 is a low cost RRIO-OPAmp. If you need highest precision you should take a more expensive OPAmp.

    Kai
  • Hello,

    As Kai confirmed, the maximum input offset voltage is specified at a Vs/2 common-mode input voltage.  The worst-case input offset voltage at the maximum common-mode voltage is indirectly specified through the CMRR specification.  The minimum CMRR is 57dB over the full input common-mode range, including the upper region where the input pair transitions.  57dB is roughly 1.4mV/V, so over the 2.75V change from 2.75V to 5.5V the maximum offset in the upper region should be less than 3.9mV.  A really worse-case calculation would take the +2mV maximum initial offset and then add the +3.8mV from CMRR.  This situation would be extremely unlikely.