TeX - LaTeX Stack Exchange is a question and answer site for users of TeX, LaTeX, ConTeXt, and related typesetting systems. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

No too much more to say.

\( 2.63\si\ohm \) looks odd. But it might be the correct way.

\( 2.64\,\si\ohm \) looks much better IMO. What about the \,? Sould it be bigger, smaller or there is no really a convention for that?


share|improve this question
up vote 35 down vote accepted

As the NIST Guide for the Use of the International System of Units (SI) states

7.2 Space between numerical value and unit symbol

In the expression for the value of a quantity, the unit symbol is placed after the numerical value and a space is left between the numerical value and the unit symbol. The only exceptions to this rule are for the unit symbols for degree, minute, and second for plane angle (...) in which case no space is left between the numerical value and the unit symbol.

Note that this is what happens if you use the siunitx as intended, namely by using \SI{<value>}{<unit>} to print a quantity:



enter image description here

share|improve this answer
Note that the quantity is formally a product of the number and the unit, so the space here is showing multiplication. – Joseph Wright Jun 17 '11 at 6:14
@Joseph: Thinking of the quantity as a product of number and unit can be quite useful (e.g. it explains why in tables and graphs, the unit should be separated from the dimension by a forward slash indicating division), but one has to be a bit careful: While different units may be separated by a raised dot to indicate the multiplicative nature, this is not permitted between the number and the units. – Jake Jun 17 '11 at 7:27
That's a typographic convention: in terms of unit analysis these are all products. – Joseph Wright Jun 17 '11 at 7:38
@JosephWright I disagree. It's simply a notational convention. Jake is right. The same way the "dx" in an integral is in ordinary analysis not multiplied with the function term (unless you reinterpret it as a differential to justify the notational abuse in retrospect), the correct way to think of a number-unit cluster is as a pair <number,dimension>. To convince you of this: no number can ever modify a dimension and no dimension can ever modify a number. When you think of "2m" denoting "two meters", this really means "2m = 2 * 1m" (= 2 * <1,m>), not "2 * m". – Lover of Structure Nov 3 '12 at 5:03
@Jake [still addressing Joseph, but I'd like Jake to see this:] There are other constraints: not only does one never write "34·m" (like Jake has pointed out), one also never writes things like "m·1kg"; it must be "1 kg m" or "1 kg·m", and the reason is that there is an underlying constraint (normally obeyed, though I'm sure there are exceptions) that no unit cluster is preceded by nothing. Technically speaking, "(3m²)/(m)" (displayed as an ordinary fraction) is notational abuse as well, because it should parse as "3 (m²)/(m)", but admittedly that is tolerated. – Lover of Structure Nov 3 '12 at 5:10

This is more of an answer to an answer (too long for a comment; no picture), but hopefully it'll serve as an illustration to the question as well:

\def\hilit{\hskip-.1pt\smash{% negative skip for the vrule width
  \special{color push rgb 1 0 0}\vrule width.1pt height4.5ex depth6ex\special{color pop}}}
  \SI{2.63}{\ohm}               \par
  $2.63\hilit\,\hilit\Omega$    \par % which is the same thing as...
  $2.63\mskip\thinmuskip\Omega$ \par % which is the same thing as...
  $2.63\mskip3mu\Omega$         \par % 1em == 18mu; mu == "math unit"
  $dx\,dy$                      \par
  $2.63\hphantom{\cdot}\Omega$  \par % we're talking about a product
  % product = "a quantity obtained by multiplying quantities together, or from
  % an analogous algebraic operation." --New Oxford American Dictionary
  $2.63\hilit~\hilit\Omega$     \par % ~ == "non-breakable space" or "tie-in"


share|improve this answer
Great comparison. I'll stick with the siunitx way for now. – Tomas Jun 18 '11 at 21:31

My preference would be \;\si\ohm rather than \,\si\ohm or \si\ohm. As Jake mentions, there is a convention, though it strikes me as ill-defined. "A space" is either literal or typographically ambiguous. Definitely some kind of space is preferable to none, but the exact size is likely up to personal taste outside of technical papers (where the behavior is defined, one hopes) or when not using the facilities provided by siunitx.

share|improve this answer
@Jake Well, that's me sorted. :D – Jack Henahan Jun 17 '11 at 4:16
you could delete the answer, then – blubb Jun 17 '11 at 13:40
@Simon Edited instead to still be correct and useful, but not to steal Jake's thunder, so to speak. – Jack Henahan Jun 17 '11 at 15:33

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.