# How to balance the vertical positions of equations inside an empheq environment

To write grouped equations, I use the empheq package with some [left=...] option , which in general works as expected.

However, when having a line with an integral sign, and I guess this may happen with other large symbols, the symmetry of the position of the equations (see the minus sign) with respect to the left hand side is lost, while the brace encompasses the whole integral symbol.

There is probably a way to avoid accounting for the integral sign, but I do not know it.

Any ideas ?

\documentclass{article}

\usepackage{empheq}

\begin{document}
% Example without integral
\begin{empheq}[left={zzzzz = \empheqlbrace}]{align}
& -xxxxxx \\
& -yyyyyy
\end{empheq}
% Example with integral on one line
\begin{empheq}[left={zzzzz = \empheqlbrace}]{align}
& -\int_{x0}^{x1}xxxxxx \\
& -yyyyyy
\end{empheq}
% Example with integral on two lines
\begin{empheq}[left={zzzzz = \empheqlbrace}]{align}
& -\int_{x0}^{x1}xxxxxx \\
& -\int_{y0}^{y1}yyyyyy
\end{empheq}

\end{document} EDIT

To make it cristal clear, I wonder if there is a way to achieve this kind of vertical positionning (sorry for the bad paint drawing). Updated to match the updated question:

Here are three possibilities:

\documentclass{article}

\usepackage{empheq,amsmath}

\begin{document}

\begin{empheq}[left={zzzzz = \empheqlbrace}]{align}
& -\textstyle\int_{x0}^{x1}xxxxxx \\
& -yyyyyy
\end{empheq}

\begin{empheq}[left={zzzzz = \empheqlbrace}]{align}
& -\int_{x0}^{x1}xxxxxx \\
& \vphantom{\int_A^A}-yyyyyy
\end{empheq}

\begin{empheq}[left={zzzzz = \empheqlbrace}]{align}
& \vphantom{X^2}\smash[t]{-\int_{x0}^{x1}}xxxxxx \\
& -yyyyyy
\end{empheq}

\end{document}


(The \smash[t] smashed it too hard, so I compensated with the \phantom{X} bit.) If the integral is in the second formula, use \smash[b] instead.

If you think the smash is too brutal, add a zero-width \vrule of a suitable height

• Thanks for the answer ! However, I don't think that \textstyle gives a symmetrical position of the two equations. I also thought of the \vphantom and I use it myself sometimes, but I think it only hides the problem. – BambOo Apr 6 '18 at 11:31
• Well, the \textstyle one is at least more symmetrical, if not completely so. As to the second one hiding the problem, then I don't think I have understood the problem. Maybe you need to explain it more clearly. Sorry about that. – Harald Hanche-Olsen Apr 6 '18 at 11:34
• I edited the question to make it clearer I hope, let me know – BambOo Apr 6 '18 at 11:57
• Thanks for your edit. That's pretty much what the result I wanted. I personally prefer the last result compared to my initial one. It may just be a matter of taste, but It think the reading is better this way. – BambOo Apr 6 '18 at 13:13