Tangent to a curve

  • Thread starter Thread starter Nick
  • Start date Start date
N

Nick

Hi,

Hope someone can help...I'm looking for a way to find a
tangent to a curve using excel. Is there a way to find the
closest point between two ranges using functions?

Thanks for any help,

Nick
 
Do you know the equation of the curve (in which case why not
differentiate it?) or do you just have selected values?

Not sure what you mean by "the closest point between two ranges"

Jerry
 
Nick

I work in a laboratory where I often have to calculate the tangent to a
curve which has been plotted from a large number of x/y points which have
come from a datalogger. I use the two excel functions "slope" and
"intercept" to get the m & c in the formula y = mx + c.

The trick is to find out what range you need to calculate it over. Too
small a range and small discrepencies in the data will cause a big error in
the slope of the tangent. Too big a range, and it may become the average
slope of a large section of the graph instead of the tangent at a point. In
the work that I do I have found ±3 % of the max y value to be a reasonable
compromise. This will be different in a different situation. I actually
make provision to be able to change the ±3 % for the odd situation where the
tangent doesn't look right. ie the data is sometimes a bit rough, or there
may be not enough data points in the ±3 % range.

If you have a formula for the line you can get the tangent very easily as
Jerry suggested, by differentiating it. In this case don't use my method!

Lets know how you go.
Alan
 
Nick

About the closest point between two ranges.
Not sure I know how to do it.

Before I try to work it out are the graph lines formulas or random test
data?
Do you want the x distance or the y distance or the actual closest distance?

Can you give me some idea of the shapes of the lines? closed loops? etc

Alan
 
Hi Alan,

I've got formula for the curved line. I plan to
differentiate after I get the correct x value.

Say the x axis is time. I need to know at what point in
time the slope of the straight line is the same as the
slope of the curve. From this I could get the intercept of
the straight line, which is really what I'm after!

I was hoping someone knew of a clever way to get "x", so I
could automate it.

Thanks,

Nick

-----Original Message-----
 
Back
Top