[Phys-l] unit conversions (was: Mass and Energy)

[John Denker <jsd@av8n.com>]

John Mallinckrodt saved me a lot of typing:

> In my opinion the consistent use of SI units in introductory physics  
> courses only exacerbates the problems students have with units,  
> problems that they WILL face in the cruel world, not to mention our  
> own introductory laboratories. (Does anyone really conduct  
> introductory laboratories in such a way that students never have to  
> face unit conversions??)  Far better to make sure that they  
> understand that units are NEVER to be taken for granted 

Amen, brother.

> and that  
> unit  conversions are ALWAYS trivial--a little tedious perhaps, but  
> always trivial.)

I think I agree with the sentiment ... but "trivial" isn't the
right word to express it.  Over the years, "trivial" has come to
mean "insignificant", whereas unit conversions are quite significant.
Perhaps routine or straightforward would be a better word.

> Students in lab are constantly asking questions like, "Do I have to  
> convert the length to centimeters?"  My standard response is  
> "Certainly not.  Do so if you want to, but units always take care of  
> themselves as long as you treat them with enough respect to let them  
> do so."  When you don't do unit conversions BEFORE a calculation you  
> may end up finding that some force is (2.3 +- 0.1) x 10^-5 W h/ft.   
> There is nothing wrong with that EXCEPT that it may be hard to  
> compare with the product of m and a that you found to be (352.5 +-  
> 0.5) g and (75 +- 2) cm/s^2 respectively.  No problem, just do  
> whatever conversions are required in order to allow the comparison.

Yes.

> The fact that problems like this arise constantly in lab should be  
> all the evidence we need to appreciate the fact that students will  
> not be shielded from the need to perform unit conversions in the real  
> world.  Thus, an important part of our job ought to be making sure  
> that they face that need squarely and with the kind of confidence  
> that can only be borne of their own appreciation that a respect for  
> units is critically important and that unit conversions are both  
> routine and trivial.

Routine and straightforward.

==============

I have little to add, except to say that I don't even know how to
ask for a cup of coffee using units that don't require conversion.
The SI unit is m^3, which is impractical.  The liter is "accepted
for use with SI" according to
  http://physics.nist.gov/Pubs/SP811/sec05.html
but requires conversion before being used with SI base units.

In any case, trying to avoid unit conversions is a losing strategy.
My advice is: when doing a long calculation, pick a consistent set
of units and do the conversions *early* in the calculation, so
that the main body of the calculation uses consistent units.