• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 49 subscribers
  • Views 183 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.

VOLT-DIVIDER-CALC: Gives wrong values and has limited capabilities

Jacob Vech
Prodigy 40 points
Part Number: VOLT-DIVIDER-CALC

It is hard to believe that a design aid as poor as "VOLT-DIVIDER-CALC" should exist on TI's otherwise excellent web site.

1) The first thing I noticed is that it often gives answers that are plain wrong. For example when I asked to divide 60 volts down to 1 volt using E24 resistors, it offers 75K and 1.3K giving a 2.2% error. However, using 33k and 560 ohms would give an error of just 0.1%. So it fails in its primary duty: to find optimal resistor pairs. 

2) It cannot offer resistors for a 100:1 ratio or more, that is out of range. Why? That is a most typical resistor ratio that one might need.

3) Its lack of capability is disappointing. For example one cannot ask it to select resistors from the superset of E24 and E96. Usually, whenever E96 resistors are available, so are E24 resistors. Almost always, an optimal divider ratio will come from combining the two ranges. For example, if I need a 30:1 voltage divider, 348K and 12K have zero ratio error, but the best that the tool can suggest has a 0.4% error.

This tool has the level of sophistication I would expect from an overnight assignment for a 6th grade beginners programming school class, yet it is offered to professional engineers as a design aid. To be useful, it should offer more than such a minimum capability. For example, engineers are often willing to add a third resistor, either in series or in parallel with one of the resistors in order to get a satisfactorily low initial error. Such a tool should offer optimal 3-resistor solutions. Another frequent need is to find three or more resistors with a given ratio, like 1:9:90 or 1:2:4:8. Another normal requirement is to be able to specify minimum or maximum acceptable resistor values. But this tool offers nothing but the barest minimum, and even that has bugs.

Maybe someone at TI could spend a day or so writing a useful tool, or at least correcting the actual bugs. Or maybe someone in this forum would like to write and post a more serious resistor selection design aid. 

It would certainly be good to have some feedback from TI themselves to the issues raised here. 

  • 0
    TI__Intellectual 2630 points

    Hi Jacob,

    Thanks for the above post. Unfortunately the post has come to WEBENCH forum,  VOLT-DIVIDER-CALC  does not belong to this forum, We are moving this post to other forum which can help in answering the above post.

    Regards

    Shaheen

  • 0
    TI__Mastermind 25795 points

    Hello Jacob,

    Thank you for your post.

    Please evaluate the Analog Engineer's Calculator. You'll find it much more capable. For voltage dividers, simply select Passive->Find Voltage Divider.

    Concerning the VOLT-DIVIDER-CALC, I am unsure of who created it. It is also approximately 21 years old. We will attempt to remove that page as it appears to no longer meet the needs of our customers.

  • 0 in reply to Pete%20Semig
    Prodigy 40 points

    I appreciate that it is agreed that VOLT-DIVIDER-CALC should be retired, and also appreciate that the bugs present in VOLT-DIVIDER-CALC do not appear in the voltage divider facility of the "Analog Engineers Calculator". However, the new tool still fails to provide a useful way to find an optimal resistors for a voltage divider for the reasons I detailed in my point (3) above. Once again, it does not allow you to select resistors from the combined set of E24 or E96 resistors, while in practice this will almost always provide the best solution, and again, it does not allow the introduction of a third resistor, which is almost always needed to achieve a close initial tolerance. And once again, it does not provide a way to find a set of more than 2 resistors, for example 1:2:4:8. In other words, despite this being a very large program (400MB download) it does not provide useful results in practice. We still do not have a tool to allow us to chose optimal resistor values in most real design situations.

  • 0 in reply to Jacob Vech
    TI__Mastermind 25795 points

    Hello Jacob,

    Can you please point me to a tool that does all the aforementioned calculations outlined above and in point #3?

    I can then feed that information back to the person who created and maintains this tool and perhaps he can include it in the next revision.

    Thanks,

    Pete