It’s Good Enough

I distinctly remember the answer to a question that my professor asked one day in grad school.  Too bad I don’t remember the question.  It was somewhere along the lines of “Why should we accept this result?”  Several students tried to answer the question with an analytical or calculated result.  But no one got it right.  The answer, according to the professor was “Because it’s good enough.”  His point was that we may never find the true answer, but the answer we got is close enough.

I am reminded of this a lot in our work.  We deal with software that calculates at a ridiculous precision level – something like 16 decimal places.  Yet we deal with disciplines like surveying and civil engineering, where sometimes ½ of an inch is an acceptable tolerance.  We often assist a construction surveyor and I can’t help but laugh sometimes at the level of precision he is okay with (see ½ of an inch above), especially when the civil engineers that created the plans were probably fretting over a lost 1/100 of a foot somewhere.

To me, this is one of the things that separates an inexperienced surveyor or civil engineer from an experienced one – understanding when the answer or result is “good enough.”

Civil 3D helps a little with putting precision in perspective; because you really can’t cheat or fudge with the software.  The labels are a result of the design.  The invert elevations shown at a sanitary sewer structure are a direct result of the pipe slopes, lengths and starting elevations that you entered.  If that invert elevation is off by 1/100 of a foot from your design, with previous software you might recalculate everything by hand (If it’s off by 1/100 of a foot, maybe it’s really hiding a egregious error!  Engineers are a paranoid sort).  But in Civil 3D, you know the pipe slopes that you entered are solid, you know that the manhole locations are solid, and you know that the starting elevation is solid.  If that invert looks off by 1/100 of a foot, it’s probably a rounding error.  Let it go.  It’s good enough!

2 Comments
  1. Something as important to go along with this is the hydraulic design of the pipeline. On a recent project, we were microtunneling and the contractor made an error and installed the pipe 6 inches too high in one reach. We are still dealing with a fix for that. That affected the slope downstream and upstream. The contractor was never closer than about and inch to grade, which was the required precision. The pipe that we are using has a 360 degree PVC lining so the Manning’s n when new is about 0.011. Standard design criteria also calls for setting design depth at 0.75D. But we knew that we would have problems with installation grade so we used an n value of 0.013 bare concrete and a d/D of 0.666. This provided assurance that after construction the pipeline would have adequate capacity to carry peak flow in the future at standard depths and still have a small additional capacity if the contractor did an excellent installation job.

    What does this have to do withthis topic? Young engineers today doing hydraulics would say that they should use a Manning’s n value of 0.011 and a d/D of 0.75. But we have to keep in mind that we do not build things in America like the Germans do, with precision. We get low bid contractors that cut a lot of corners so we need to keep that in mind and use conservative design criteria so that our clients get the capacity they deserve on underground work. In the future it will be more and more difficult because our streets are getting clogged with utilities. Be conservative and remember that “an estimate is good enough” but be careful with the design criteria.

    Nice article.

    R

  2. Drew:
    Your article is on point. Gradestaked for various projects the question “Is it Good Enough” always came up. (ref, worked with you on Oakland project 10 years ago).

Leave a Reply

Your email address will not be published.

Back to blog

Copyright ©2017, CAD Masters, Inc.. All Rights Reserved.