# Align and sub-indexes.

Consider

\begin{align}
u_j &= x\\
v &= x
\end{align}


Compiling this the $=$'s are going to be aligned, but the $v$ gets aligned to the end of the $u_j$, which is a bit beyond the end of the $u$ because this has a sub-index.

Question: Is there a robust way of correcting this?

• v\phantom{_j} &= x. A related question: tex.stackexchange.com/questions/111722/… – Steven B. Segletes Mar 26 '14 at 19:14
• I intended to mean by robust something that is not a tweak like this. But seen this solution made me think that there is no way for latex to know the extra space is coming from a sub-index or not. I should have though the question better. – Anna Taurogenireva Mar 26 '14 at 19:21
• Oh, the link is interesting. – Anna Taurogenireva Mar 26 '14 at 19:22
• The tabstackengine package can do it without a tweak: \documentclass{article} \usepackage{tabstackengine} \stackMath \begin{document} $$\TABbinaryLeft \tabbedCenterstack[l]{ u_j &= x\\ v &= x }$$ \end{document}. However, it can't (yet) number each equation separately, as does align. Also, there are some configuration commands as overhead, so unless you were to use it often, the tweak is easier for your example. – Steven B. Segletes Mar 26 '14 at 19:30

If you don't want to use the hphantom{} approach (which I would consider better in this specific case), then you can use the alignat environment which provides pairs of r/l alignment points:

## Notes:

• The first & makes the content following left aligned. Then you need to skip past the subsequent right aligned column (hence the &&) to get the x to be left aligned.

## Code:

\documentclass{article}
\usepackage{amsmath}

\begin{document}
\begin{alignat}{3}
&u_j &&= x\\
&v   &&= x
\end{alignat}
\end{document}