• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 48 subscribers
  • Views 712 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.

LMP7721 Bias-current

Ishara Mudalige
Intellectual 280 points
Other Parts Discussed in Thread: LMP7721, TINA-TI, IVC102, LMC6001

Hi,

This is a follow-up question of my previous post

https://e2e.ti.com/support/amplifiers/precision_amplifiers/f/14/t/425048

Here I have attached a Tina-TI simulation file of LMP7721 bias-current simulation. The calculation shows 20fA bias current (at the inverting node). Is this simulation right?  I’m wondering, why I can’t get down to 3fA of this simulations?

Thanks

~IsharaLMP7721_bias_current.TSC

  • 0
    TI__Mastermind 49956 points
    Hi Ishara,

    There is a note in the model that you must change some SPICE settings to resolve the femtoamp currents. The defaults do not have enough resolution. Gmin sets the maximum open circuit resistance of the switch (1/gmin)

    * NOTE
    * USE ANALYSIS OPTION GMIN OF 1E-13 TO 1E-15 AND
    * RELTOL = 0.00001 TO 0.000001 FOR ACCURATE INPUT
    * BIAS CURRENT

    These settings can be found under ANALYSIS -> SET ANALYSIS PARAMETERS.

    Also note that the common mode is at the negative rail. Bias currents will be higher at the rail than at mid-common mode.

    I could not duplicate your waveform, but give it a try.

    Regards,
  • 0 in reply to Paul Grohe
    Intellectual 280 points

    Hi Paul,

    Thanks for the reply. I changed the SPICE settings accordingly and resolve the problem. However, I couldn't find the settings to change the bias current. Is it same as the "DC absolute base current error"? 

    By the way, I have some other question.  

    I have seen a video " what's all this Femtoampere stuff, Anyhow?" of you and the late Bob Pease were in. Just wondering, the video was produced in 2007 and it seems circuit is dated back to 1993 July. is there any circuit improvements on precision femtoampere integrator circuit? For an example, IVC102 is a switched integrator TI amp, Can we use IVC102 on that circuit? ( original design has external relay drives, aren't those adding noise to the circuit?)

    Thanks

    Ishara

     

  • 0 in reply to Ishara Mudalige
    TI__Mastermind 49956 points

    Hi Ishara,

    Yes. "DC absolute Current Error". Set it lower than 1e-15.

    I built the "gold box on rubber-bands" around 1991-1992 to investigate ways to do production tests for the LMC6001 (tested to 25fA in production).

    I do not believe the IVC102 was around back then...but it would have been helpful!

    We used Coto reed relays because they have minimal charge injection (the guard surrounds the entire reed). The problem with JFET/MOSFET switches is the charge injection caused by the gate capacitance.

    It is possible to "balance" and compensate the charge with multiple switches - but that becomes complicated and adds more possible error sources. A reed relay is much simpler.. Space was not an issue, so relays are fine.

    But, yes, now you could replace that system with the IVC102. I have not used it...but I'm sure they added charge compensation to reduce the "jumps".

    Regards,

  • 0 in reply to Paul Grohe
    Intellectual 280 points

    Hi Paul,

    I am not sure , I understood this term of your reply correctly, " Also note that the common mode is at the negative rail. Bias currents will be higher at the rail than at mid-common mode" .

    I really appreciate if you could elaborate that for me. 

    Thanks

    Ishara