• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 49 subscribers
  • Views 221 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.

TLV9062: Slew Rate vs Vs

Kazuhiko Hirano
Genius 11643 points
Part Number: TLV9062
Other Parts Discussed in Thread: TLV9052

Hello,

 

My customer needs >5V/us slew rate with TLV9062 when Vs=3.3V.

Datasheet says the slew rate (typ) is 6.5V/us @Vs=5V.

My customer is concerned that the slew rate might be proportional to power supply voltage.

For example, the slew rate @Vcc=3.3V is 3.3/5 times of the one @Vs=5V, this means it is 6.5V/us x 3.3/5 = 4.29V/us < 5V/us.

Is this customer’s concern correct?

 

Best regards,

 

K.Hirano

  • 0
    Guru 47545 points

    Morning Hirano-san, 

    Actually quite a number of questions there - 

    1. How has he decided he needs a min SR = 5V/usec. Couple of different approaches depending on application

    2. The specified slew rate at gain of 1 is likely worst case and input limited, if operating at higher gain, the SR could actually be better while the BW lower. 

    3. This is only a typ slew rate, you can also infer a range on slew rates from the supply current range. They only give a max from the typ that is 40% higher - that seems extreme, but worst case maybe a 40% lower value implies a 25C production range on SR of min 0.6*6.5V/usec or 3.9V/usec. 

    4. And yes, if the supply current goes down or the comp cap goes up with the supply going to 3.3V, you should expect lower slew rate. 

    Frankly, perhaps you should start out with a device nominally in the 10V/usec region typ. 

  • Former Member
    0 Former Member

    Hello Hirano-san,

    Michael has made many good points here.  It is important to consider that this is only a typical specification and that operating too close to the specification may be an unacceptable risk for the customer.  A good indicator for a change in slew rate is the quiescent current vs. supply or the quiescent current vs. temperature.  This is because slewing occurs when all the quiescent current in the input stage is pushed to one of the two amplifier input legs and to the internal compensation capacitor.

    For this part, the quiescent current is rather stable across supply voltage and temperature.  So, I would not expect the slew rate to change too much.  However, process variation (device to device) might be considerable enough to make this too close to the design limits.

    Here is a really good e2e post on this subject from 2013.  I recommend you read the comments by Marek Lis.  Regarding the effect of gain on slew rate, I would say this.  Increasing the gain will not change the slew rate of the part itself.  This will be constant.  However, increasing the gain will decrease the difference between the inputs for a large step voltage.  This means that the amplifier will spend less time in the slew rate mode.  Hopefully, this idea is clear.

    If you decide the customer needs a part with slightly more slew rate, I would recommend the TLV9052.  It is closely related to the TLV9062 device, but with more slew rate.

    Regards,

    Daniel

  • 0 in reply to Former Member
    Guru 47545 points

    Hey Daniel, 

    I wouldn't discount that gain of 1 comment so quickly - essentially, the point is what is the required input CM slew rate to follow the input. There is very little bench testing of that these days, but I have done some over the years where, for instance, CFA amps have a unity gain buffer across the inputs - it has its own slew limit that is part of the design flow. At a gain of +2 definition point, we just need it to be more that 1/2 the output stage slew rate as the CM swing is 1/2. Those parts, run gain of +1 will degrade on slew rate vs the spec gain of +2 due to the input buffer. Similarly, running a CFA inverting (no CMswing) will always give a higher SR than non-inverting as you bypass the buffer CM slew rate. 

    I would suspect if anyone tested it, the SR for these low power VFA's is incrementally higher inverting gain of 1 than the typ spec. of +1 with no input CM swing. And at higher non-inverting gains with lower input CM swing, the may also go up on slew rate a bit. 

    What I was implying was, if you really wanted to drill down to the available slew rate on a part, you kind of need the external circuit - however, I recognize even with that you are probably back to bench testing as virtually no parts provide this level of detail - hence, use something with more design margin is what you are left with lacking data basically. 

  • Former Member
    0 Former Member in reply to Michael Steffes

    Hi Michael,

    Thanks for the input and taking the time to explain based on some of your past experience.  Much appreciated.

    I think for this customer, as you have said, the best way forward may just be to leave a measure of margin in the design.  We will see if they have any follow up.

    Thanks,

    Daniel