Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] frequency: a modest proposal

On 01/27/2010 01:33 AM, James McLean wrote in part:

... why not define the unit 'cycle' = 2pi radians, and then make
1 Hz = 1 cycle/s = 2pi radians/s = 2pi/s ?

That makes sense ... and is indeed the only definition
I have ever used, or ever heard of, until now.

I've always been mildly disturbed by the definition of the unit hertz as
equivalent to 1/second. (Such as can be found, for example, at
<> and

I'm surprised. I've never noticed that before. It
looks like a bug to me.

Can anyone see any problems with this [radian] option, other than the near
impossibility of changing tradition?

It's not a problem. It's not even a change. Radians
and radians per second (not cycles per second) are
already traditional throughout mathematics, throughout
electrical engineering, and in every physics book I
can think of.

I cannot imagine any argument in favor of cycles/sec
as equivalent to 1/sec. I have to assume that those
two NIST pages are just mistakes.

Suggestion: Send a short note to the NIST guys and
suggest they repair the web page to show Hz as
2π radians per second? Or call 'em on the phone.

I would hope a very short note would suffice. That
is, I hope it is not necessary to argue the point.
It should suffice to say that 2π radians per second
is consistent with the other units on the page, and
s^-1 is not consistent. I cannot imagine there would
be any sort of counterargument.

You could start with Thomas O'Brian, chief of the
Time and Frequency division. Whether or not he's
the exact right guy to handle this, it's his job to
know who the right guy is.