• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 48 subscribers
  • Views 267 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.

LMV934-N: Unused elements left completely floating

Dean Weiten
Prodigy 130 points
Part Number: LMV934-N

Hi,

I have a case where a new configuration of an existing product requires a depopulation of some channels of an analog circuit.  As a result, some unused element(s) of an existing LM934 are left floating.  I can re-populate a couple of components to tie the non-inverting input to a known potential, and put a capacitor across the two inputs as well.  I suppose I could even replace the capacitor across the inputs with a resistor, bringing the two inputs to the same potential.

What is the best course of action in this case?

Could the completely-floating unused element cause reliability problems in the other elements, over time?

Thanks.

Dean.

  • 0
    Guru 140200 points

    Hi Dean,

    take a resistive voltage divider from the supply voltage to signal ground to produce a potential at midd-supply and connect it to the +input of free OPAmp. Connect the -input to the output of OPAmp to form a voltage follower.

    Kai

  • 0 in reply to kai klaas69
    Prodigy 130 points

    Yes, if I had design control.  The board exists and we are loathe to change it, would be expensive and cost a shipment delay.  I can tie one side to a known potential, and put component across the inputs, but I can't add PCB nets at this time.

  • 0 in reply to Dean Weiten
    TI__Mastermind 33011 points

    Dean,

    Please see the following blog about what to do with an unused op amp. Please also consider the devices common mode voltage range in your changes. LMV934 is RRIO, but not across all temps. So you may be able to tie it to ground, but the best practice is to connect it in buffer configuration.

    https://e2e.ti.com/blogs_/archives/b/thesignal/posts/the-unused-op-amp-what-to-do 

    Best,
    Jerry

  • 0 in reply to Dean Weiten
    Guru 140200 points

    Hi Dean,

    hhm, then I would try connect the +input to ground and mount a 1M resistor between the inputs. The input bias current of -input will cause a voltage drop across the 1M resistor which should be higher than the input offset voltage. By this measure wild output voltage togglings (or even oscillation) should be prevented. No guarantee, though, from my side...

    Kai

  • 0 in reply to Gerasimos Madalvanos
    Prodigy 130 points

    Hi Jerry,

    Yes, thanks, I had tripped on that blog article based on another question :-)  It aligns well with my intentions on all of my own designs (40 yrs plus now).  I like the paragraph that says something like "if you aren't perfect, don't sweat it, most modern op amps are well behaved, just please do better next time."  Sage advice, and makes everyone feel better :-)

    Dean.

  • 0 in reply to kai klaas69
    Prodigy 130 points

    Hi Kai,

    Yes this is similar to what we decided to do after our deliberations.  We don't have access to -Vs in the circuit, but we can peg it to a voltage that's within the common mode range, and then put an impedance between the inputs.  It turns out, with some other digging, that this approach has been used in a similar application on another one of our products with the same issue (depop of a stage that's unused in a particular configuration).  Based on all this, we are feeling more confident in that approach.

    I will try to ensure that future implementations allow for better termination of unused elements, in all circumstances

    Best regards,

    Dean.